Everpure Stock

We are excited about today's discussion and the future of Everpure. Let me remind you that we will be making forward-looking statements today that are subject to assumptions, risks and uncertainties. Actual results could differ materially from those anticipated due to a number of factors, including those referenced in the detailed disclaimer at the beginning of our presentation slide deck and in our public filings with the SEC, which we encourage you to review. The presentation slides discussed today will be available on our Investor Relations website at investor.everpuredata.com. Today's discussion will include non-GAAP financial measures. Reconciliations of these non-GAAP measures to the most directly comparable GAAP measures are available on our Investor Relations website at investor.everpuredata.com.

Welcome back, everyone. Good afternoon, good afternoon to everyone that's on the webcast. And I hope you enjoyed the lunch. I know it was a bit early, but for those that were joining from the East Coast, perhaps welcome time to have lunch. So we're looking forward to speaking to you this afternoon. I'm just going to briefly go into the agenda again for this afternoon. One is I'm going to present to you a very high-level overview of our overall strategy, reminding you where we've come from, but also how this fits, where we're going and how that fits into the context of Pure Storage and our overall business strategy.

Rob, you'll see Rob again coming on stage talking about how these different areas in which we're competing, why we have a right to win there, what underpins our technology, and I'll be going into a little bit more detail on the hyperscale side of the business and reason why we are as compelling a solution as we are in the hyperscale. And then after the market closes, we'll have Tarek come on stage and tell you what that means in terms of our financial profile for the next several years. And I'm sure that will be something that you're very much looking forward to, but we're going to wait until after the market closes to get into that. So with no further ado, I mean, the story is actually quite straightforward.

Have you ever seen a line that straight in any financial environment before? Look at that line. It's about as straight as it can be. This is our increase in market share year after year, okay, in all kinds of markets, right? This is well over 12 years now of share gains each and every year. right? We outpaced the market and we outpace our competitors. And this -- it's not just year after year, it's quarter after quarter. It's a very remarkable and consistent share gain. So when we have said on earnings calls that we have 2 quarters of visibility, that's actually quite true from the standpoint of pipeline in Salesforce and from our sales team. But we have 12 years of visibility of continuous market share gains. And that gives us a lot of confidence that we're going to continue to see market share gains as we go forward. In fact, we believe that's going to accelerate.

If you want to know what is the most fundamental element of our strategic competitive advantage in this space, it is this. We treat data as high technology, not as a commodity, all right? Why is that a differentiator? Well, if you were to ask Joan Magretta, who's up there with Michael Porter in terms of one of the foremost business strategy analysts, they will tell you that if you structure your business, which is then displayed in a P&L, if you structure your business around a different and compelling value proposition. That, that can be sustainable because it's very difficult for your competitors to mimic that business model. We have a technology-forward R&D-led business model, high margin, but also high investment in R&D. We are a technology-first company, and most of our competitors are commodity players. They cannot change -- they cannot overnight or even over many years, become a technology player. Think of what that would do to their P&L in the medium term. This is a fundamental structural advantage that we have as we go forward, and we continue to invest in innovation. And in fact, if you look at the last 12 months, on a GAAP basis, it's now over $1 billion a year, and this is, of course, increasing and that puts us at the #1 investor in R&D in our segment, okay? So it's been true for many years that we spent more on a percentage of revenue, but now we actually spend more dollars. And I would argue that we spend it very efficiently and effectively in our business. And that's allowing us to gain market share.

What does it buy us? Well, at least third parties, whether it's Forrester or Gartner, have consistently put us as the innovator in the space, always in the top quadrant. And at least for the last many years, the furthest north and the furthest to the right, and so we are being recognized increasingly not just by analysts, but more importantly, by customers as being the innovation champion in this space. And we're not just the innovation champion, but if you look at our Net Promoter Score, nobody even comes close, right? So we're very, very customer focused as well. And now we have scale to back this up. And it's not just that we invest more, but we're more efficient in that investment. You heard all morning about our unified data plane, about a unified control plane about Evergreen. So what are the advantages of this?

Well, having 1 software environment that covers many different markets is the very definition of scale. Because we invest once and we can serve many different markets with the same investment. We don't have to invest in a 1,000 different operating systems to service different parts of the data storage market, right? This one investment we make in purity and in direct flash, which is part of purity, allows us to satisfy core, as you heard this morning, as well as scale AI for the neoclouds and as well as hyperscale solutions. Now these are not just benefits to us, they're benefits to the customer in terms of they get a more consistent solution overall. They get higher quality because the great thing about having a single software environment is what -- what I phrase is bug consistency. If you fix a bug once, it's fixed everywhere. You don't have to fix it multiple times. And simplicity. It is simply easier to be able to be able to manage an environment that's consistent rather than different.

It benefits the company as we said, because it is about scale and efficiency of investment. It improves our overall quality, of course, and the focus of the team overall, right, whether it's the sales team or the engineering team. And then what are the shareholder benefits? Well, really, we get efficiencies in scale and growth, right? And over time, that should pay off in terms of enhanced profitability overall. So we spend more, and we are more efficient in the way we spend it. Over the years, we've also put a lot of investment in our go-to-market team, right? And if you look at it now, we have over 2,300 professionals in our go-to-market organization. They are part of that Net Promoter Score that I mentioned. It's not just about product. It's about the way we engage with our customers, right?

When we speak about Net Promoter Score, it really is a measurement of every aspect of our company whether it's the way we engage with our customers, the way our channels engage with customers, they are part of this measurement. The way we interact with our customers from a financial standpoint because that's part of the measurement. And of course, the quality of the products, the quality of the support that we provide those customers. In the enterprise, we're now -- we're in 64% of the Fortune 500, 44% of the Global 2000. We have a lot of global strategic partners. By this, we mean technology -- well, both technology partners at 50 global strategic partners in terms of customers we go to market with -- sorry, partners we go to market with. We have 4,000 channels worldwide in the commercial market. We're widely respected by our channel -- by our channel as being a company that works with the channel well, a company that is trusted by the channel that provides the channel what they need to be successful, consistently ranked #1 by CRN and many of our channel partners we have a direct presence in 32 countries.

We sell in well over 60 countries on an international basis, 15,000 customers overall. And increasing our presence in governments on a worldwide basis. That had been a weakness of ours in the past, and now that's an area of growth for us. All right. And that's allowed us to reach an inflection point. So if we look at it solely from a reported revenue basis, now up 38% year-over-year. I'll call your attention to the fact that we have 8 year -- 8 quarters of increasing revenue growth. okay? So it's not just about the most recent price increases, okay? We've been increasing our revenue growth consistently over these 8 quarters, gaining share. And this is why we now believe it's durable. This is not just a spike.

And we're growing the margin at the same time. All right, a 50% increase in the overall margin of the business over that same period of time. And again, this is, we think, proof that we drive growth through innovation. It's innovation-led growth. So we're going to -- I'm going to very briefly speak to you now about the 4 areas that we believe we are engaged in as we go forward. Of course, we've been in core and now core AI for almost 2 decades. Let's see if -- okay, that seems to be working better. Almost 2 decades. And that area, you might want to get me another mic. Certainty. In that area, we believe, is accelerating. Our growth in that area is accelerating. Second, we're going into, as you heard this morning, a number of different areas of software in the enterprise environment in order to enhance the data estate in that enterprise. Everything from data intelligence, which is characterizing in the data, to data stream, which is vectorizing the data to modern virtualization, which is allowing new areas of virtualization to operate at greater efficiency inside the organization.

Third, scale AI, so expanding our product area into the largest of the large AI cloud environments, 10 to 100 times higher speed performance and capacity than we had been selling into the enterprise. And then finally, last and certainly not least, the hyperscale solutions. So I'm going to give you a brief overview of these, and then we're going to follow on with Rob, who will go into much greater detail here. All right. So starting with the core, this is another look at that 38% revenue growth. But the difference here is I'm looking specifically at the product growth all right? So sales of product over those -- over the last 7 or 8 quarters. And then tied into that, the growth of our subscription business.

This is Evergreen//One, right, when we don't sell in array, but rather provide a storage service based on our arrays, right? And it's a little bit difficult to be able to identify specifically when we sell an Evergreen//One contract exactly what would that have been? Had we sold an array rather than just a service contract? Because obviously, that revenue is spread out typically over about a 3.5-year period. There's a variety of different term links that we engage with our customers. But it's in the year 3 to 3.5 year range. So it's just a rubric. It's not -- you cannot square the circle of converting from one to the other. But very loosely speaking, if we took 70% of our TCV sales of Evergreen//One and add that to our product sales, this is the growth rates that you will see over the last 7 quarters, okay?

This is why we believe we are really accelerating in this core AI market. It is accelerating product growth at a significant level. We are outpacing both the market as well as our competitors. And we believe that this momentum is really something that is going to continue. At the same time now, we're seeing an acceleration of sales into AI environments. And I want to describe a couple of things. One is you'll notice that this is not revenue. This is GPU attached systems. So we've completely pulled out any association with price rises in this, right? This is the number of systems that we are selling into direct attached GPU environments. Not what systems might be used by AI somewhere in the enterprise. These are things that are -- these are arrays that are attached directly to GPUs and why the sudden gain really at the beginning of this year? Well, it had to do with the fact that the models just got so much better. I mean it really wasn't specifically anything we had done. Of course, we had kept developing our product. But unless the models got to the point where enterprises were starting to build our own AI factories, then we wouldn't sell into it. We wouldn't have a GPU connected device if the customer didn't have GPUs.

The beginning of this year really marked the period when customers said, these open weight models largely are good enough where we can actually get good work done with positive ROI by bringing it in-house. And once they started bringing it in-house, of course, they needed storage systems to attach it to. And so that's what we -- that's what we started to see at the beginning of this year, and it's just getting started. But I think it's -- we think that this is an area that's very much going to improve. So it's accelerating on-prem GPU deployments. And the second thing, and this is what's really critical is the combination of S and EXA proved to be a very compelling story for our customers.

There was a time where customers were being told that, well, you may not have the requirements today, but at some point, you're going to want to scale. And when you want to scale, do you want to scale on a product in an environment where you may not be able to scale and then you'd have to buy something entirely different? Or do you want to start with something that's too big for you now, but you could scale into? We're not forcing our customers to make that choice. The S and the EXA are the same technology. They're the same -- they are fundamentally the same product just structured in a different way. And with our track record of nondisruptive upgrades, with our track record of taking our customers along the journey of scaling their environment with as little disruption as humanly possible. They trust us in this area. And so the strategy of having both S and EXA in the same environment, very, very powerful.

And what we're also seeing is pilots now starting to go into production because there had been pilots before, but it was largely for -- with very large customers. They wanted to try a GPU environment but it wasn't going into production because it just wasn't proving out the ROI. Now it's going into production. And what is the difference between a pilot and production? Pilot can be down for several weeks. So you could buy a pilot and it's kind of fun to start putting it together for the first several weeks. And if it takes that long to build, okay, fine. Production is like you expect to buy it and put it into production right away, and you expect it to work, and you expect it to work with quality and reliability. And those are all the things that Rob and others spoke about this morning.

And again, this is what we've designed for. It is what customers expect from us. And so this area has been accelerating quite dramatically over the last, I'd say, 6 months.

All right. So the thing that you heard a lot about this morning, the thing that we are most excited about, though, is a fundamental change, we believe, in the architecture of IT environments that as Rob and Prakash and Ashish spoke about this morning, we think is a 10-year journey for customers. And that is this inversion of the relationship between data and applications to where data -- applications have been the way we have designed our IT environments for 30 years. Completely application dominated to the point where organizations are now fragmented in all of their applications. And it's hard to put together any 1 element now from a single application. It always requires data from multiple applications to get this going.

And the problem is not so much that there's no single source of truth. The problem is there are so many sources of truth and they just don't agree. Every one of the databases, and SAP database or ServiceNow database, a Salesforce database, your systems of operation that have databases. They all have their version of truth. And you try to figure out what is the full relationship with any individual customer and you have a very hard time. And it was already starting to break with so many applications. But now with AI that requires access to data Well, if you get different data from different applications, how does an AI know what's true or what's right? Well, you could spend a lot of tokens to figure that out every single time or you can start building technology that allows for a single source of truth or that provides context to all these different databases that explains the source of truth.

So we have put together a paper. Those of you in the room have it in front of you, those of you online, this is a QR code that will allow you to get access to it that really lays this out. We think this is a seminal paper on IT architecture, not specifically on AI, but what AI now has exposed as the flaws in current IT architecture and where we think that it needs to go. And it's going from data processing through the full client server multi-app environment and to what we believe is data primacy, that will really allow organizations to make the best use of their -- of the future application environment as well as the ability of allowing AIs, AI agents to run more of their business.

Because there's a big difference between AI in a general environment, and AI in an enterprise environment. And I can explain that very easily. I think each one of us have used AI in our personal lives, right? I know that my wife -- actually, this occurred when ChatGPT 3.0 first came out, and she took about 20 hours of work to plan out a family holiday, right? And she had all these things planned out and she asked me about it. I said, let me try something so I got I was on my iPad, and I went to ChatGPT and within about 5 minutes with about 5 prompts, I had recreated her 20 hours of work with an output of ChatGPT. And of course, as you might imagine, it wasn't exactly what we wanted, but 95% correct, is a really good answer for something like that. 95% correct. How good is that on an invoice? How good is that on financial report? I think we'd go to jail if we had 95% financial reporting.

Enterprises require 100% accurate information, 99.9% of the time. That's the difference between AI being useful as a personal productivity tool and AI being useful in an enterprise environment. Enterprises require repeatability and consistency, diagnosability, because we need to know why did I get this answer if it was not the correct answer and accuracy in their results. And that's a very different problem if your data is fragmented. If you have different sources of truth that have different information how is an AI to know? Very, very difficult. So organizations have to get control of their data. And data primacy, we can first step is data intelligence, which is, first of all, being able to catalog all your data from every source, any source in the organization characterize that data, understand the semantics, be able to provide context around it and then understand the similarities and differences between those data points in different data sets across your organization, and then be able to express that, whether it's to an AI or even to just another application environment, right? Have the same context, have the same information for the same purpose inside the enterprise.

And so we believe this is the first step to help enterprises on their way towards data primacy or much more simply just to get better answers whether it's from their existing applications or from AI agents as they go forward. So discover, classify, contextualize it and have it be explicit, not buried in an application structure somewhere.

All right. And then we're going from AI in the enterprise to AI in the large-scale neocloud environment. Rob went through this with quite -- with a lot of detail this morning, but we're seeing growing sales, for example. And as we go from an environment of tinkers and science experiments, to a world of production and reliability and efficiency and profitability we're seeing that our existing bona fides, if you will, around those elements are becoming much more compelling to our customers.

Rob mentioned this morning that we had a meeting with a customer on a Friday afternoon, did a POC and a demo on a Thursday afternoon did a POC in a demo on Friday, had it up and running on Monday. This just doesn't occur with anybody else. Just doesn't. We design products that have finished quality associated with them, not science experiments. And again, it's the same purity engine. It's the same architecture, same management capability. And by the way, as he also mentioned, neoclouds have lots of different storage requirements. Everything from the world's highest performance data engines all the way down to archive and to backup, and we can provide all of those capabilities to those customers.

And the other thing we pride ourselves on. And you really need to look into competitive offerings to really understand this, is we scale without surprises. When we publish a number, that number works under all scenarios. It's not under carefully selected environments where you have to tweak all the parameters and have it be structured in just such a way, our numbers when we -- as published, we guarantee to work under all conditions. So we scale without any surprises with our new tweaks without nerd knobs, without a lot of fussing around.

And then our hyperscale solutions. So I want to make this very clear, a very and we'll -- Rob will be going into much greater detail. But our strategy is very simple. We want to be able to replace all of their SSD and hard disk architectures with DirectFlash. That is our goal, very simple. And with that 1 solution, be able to cover their full breadth of requirements from AI levels of performance all the way down to archived levels of cost with 1 architecture and 1 structure. Why would a customer want this? Well, our reliability is 10x -- well, in the case of a hyperscaler, probably more like 5x better. Our performance is much higher and much more consistent, we have much greater density and depending on whether you're measuring against SSDs or hard disks, significantly greater density. It's from an ongoing engagement standpoint with the hyperscalers. It's a much simpler solution because it operates above the operating system, operates in user space. Does not require them to constantly tweak for every different type of storage or for every new generation that comes on board, we have double the lifetime of any other solution they have out there.

It lowers their overall operating cost. It's a less complex integration over time. And the way we're able to do this is we provide an abstraction layer between the NAND, which is the underlying media and their operating environment. And we are -- the opportunity here is huge. It's well over 1,000 exabytes on an annual basis. You pick your number. There's a lot of forecasts on that currently. We're seeing and we have contracts going forward. So we now have reasonably good visibility into this. And this is something that Tarek will be going into in a much greater detail. But we're seeing the uplift start. And of course, we have the opportunity not only with more hyperscalers, but just as importantly, if not more importantly, the expansion within existing hyperscale customers.

So I'm just going to sum up now before handing over to Rob. We believe we are transitioning the company to durable and accelerated growth. That is we're going to be growing faster, and we think that, that is durable. 8 quarter have shown that we've seen higher revenue growth and higher share gains across the board. We have -- that investment that we make, we think, is more efficient than other players out there because we can spread it across more markets and more customers. The new markets we're adding. So this is the modern data, scale AI and hyperscaler are not overlapping with our core market, right? So this is truly additional new sources of revenue and income for us. And I would say that whereas where we were 2 or 3 years ago, where we were predicting that we'd sell into the hyperscaler but didn't have one to show for it. We're now in 2, we believe, we'll be in others as well. And we've been actually tested in both SSD and HDD environments.

Now of course, with the price rise, perhaps HDD replacement is going to wait until another day. But again, it's the same solution. It's just a matter of when prices come back down. And we can operate in hyperscalers, if the latest price increases in NAND have proven anything is that we're resilient, whether it's about HDD replacement or SDD replacement, right? This is 1 solution that covers all of their storage needs outside of tape.

So with that, as you can see, we're very proud of the fact that we have 1 architecture now, but many, many paths to growth. We have our core, we have our new markets, and that's what we're going to be spending the rest of the afternoon on. So Rob? Welcome to stage.

All right. Well, welcome back from lunch. I'm going to pick right up where Charlie left off, unpacking these 4 high-growth market areas that we're focused on as we look forward. As I -- before I do, perhaps I should say that Charlie had a slide up there that said we're at an inflection point. I've been at the company, this is my 13th year. And I would say it's -- I've never been in this position where we're standing on the precipice looking forward at growth, not just in the things that we do today. But 3 really exciting new areas really just starting to take off in the industry.

So I'm going to dive into each of these in my discussion today. Before I do, I think I want to put a little bit more definition and structure to how we're thinking about how we're defining these new market areas. I'll start with the core. So we look at this as a lot of what we've been doing since day 1, serving the enterprise, serving enterprise storage systems. We've added to that. We've included in that, what we're describing as Core AI, right? So effectively, this is what we spoke about. Chad spoke about this morning, Charlie spoke about a little bit right before me. The enterprise deployment of AI systems within their environments. We're including this within the core business because it's the same customer segment. It's largely the same solution set. But also because it's different in apart. It's separate and apart from a new opportunity we've created, which is our scale opportunity, which I'll get to.

Now as we look beyond the core, I would say, extended by core AI. We see 3 new markets, high-growth areas that we're really turning our focus to. Number one, what we're calling modern data software. You've heard elements of this morning and Charlie's intro. And this is really, I would say, stepping back from it, looking at where we sit in the industry at the beginning of, I think, was going to be a 10-, 15-, 20-year long secular transition in how software is built and how application environments are assembled. And as a result, how the infrastructure that has to serve them comes into place. What are the drivers of this? Well, you've got a number of things, you've got AI, you've got new approaches to analyzing data, you've got things like open source. We'll get into all of that.

One of the things I want to highlight here is that while we're serving a lot of the same customers that we would serve in the core business, the enterprise. The way that we're going about this, delivering these solutions as software, we're doing -- we're delivering the software value above -- over and above our own storage footprint in arrays. Our goal here is to add value to customers' data footprint, no matter where that data sits, whether it's on our arrays today, hopefully our arrays tomorrow or competitor environments or SaaS or cloud environments, right? The future for us in this category is really to start adding value over and above that storage infrastructure tier.

Coming back to storage infrastructure to exciting new markets, again, very high growth that we're focused on. I would say, extending our core IP and packaging and delivering new ways to meet the needs of these newer markets. One, scale AI, what we're terming the neoclouds, the frontier model builders, the tech titans, essentially think the folks pushing the boundaries of what's possible with AI. So extending our IP to go meet the needs of this new market. And then certainly, the hyperscale solutions, which we've spoken quite a bit about, and I'll dive deeper into.

So if we step back from it, 4 high-growth areas that we're going to focus on. The last thing I'd emphasize on this slide is that we're approaching these areas not as disparate efforts within the company. We're approaching this -- each of these markets based on a single solid foundation. That solid foundation is built on a shared set of IP that we've developed over 17 years, perfected in the enterprise are now finding ways to go repurpose, repackage that IP to meet the new market needs. And we're doing this on the basis of a significant customer base and a significant understanding of what the enterprise needs.

All right. So let's start with the core, business core plus Core AI. Equation here is fairly simple, right? We're going to keep gaining share. And we're doing it at a point in time when the market is growing. That market is growing for a number of reasons, but I would point to AI. I think you've got 2 effects driving core enterprise storage market growth from AI. You've got an indirect effect and a direct effect. The indirect effect, fairly simple. AI makes data more valuable, how much data does an enterprise store. It depends on how much value they get if they get more value, they got some more data.

The direct effect, this is the deployment of new AI attached use cases, things like as inference comes back on-prem, within enterprises digital walls, agent deployment in real-time systems. We're starting to see these projects go from pilot to production. And so we see that growing the enterprise market as well. But really 2 legs to this. We're going to gain share, we're going to keep gaining share and the market is growing.

So what are the drivers for our continued share growth? Charlie showed the straightest financial chart in history. We're going to keep driving that. Actually, if you look at that chart at the very end, there's a little uptick up, we're going to keep driving that same level of growth, if not even faster. And this is really based on a couple of key advantages that we've assembled over the last 15 to 17 years. Number 1 is we now are able to approach this market with an entirely complete platform and what I describe as an enterprise-wide value. If you go back 6, 7 years in the company's history and you look at a lot of our enterprise sales approaches, it might look something like this. You go to the customer, say, "Hey, Mr. Customer, you've got a database. I've got the best thing for you. I've got the fastest array. It's going to save you so much time. Your DBA is going to be so happy, and we nail it out of the park."

The problem with that is you only get so far as a point solution provider, you're not seen as a strategic vendor, you're not seen as a strategic partner. Well, we're now in a completely different position, having completed the portfolio, having been able to articulate and deliver that portfolio, not just with the unified data plan, but now with the intelligent control plane, we can go after these enterprise accounts with an enterprise-wide value, a franchise value, right? We talked about we quantified this in the 2Q call. When we think about large -- there was a point in time in the company's history where if there was $1 million, $5 million deal, like I would probably know about it. We're doing 8-, 9-figure deals these days, and I see them as they go by, right? And so this is just a testament to the fact that we're now in a position to go compete and land very, very large deals in the largest enterprises.

Not just land, but expand. We're in a much better position to expand. And I would call out 2 things here. One is we have the stickiest product in the market. And the second is network effect. When I think about stickiness, Evergreen comes front and center. Chad did a great job talking about the difference in mindset when it comes to the traditional rip and replace motion that customers go through versus an Evergreen modernization. No customer goes through a nondisruptive Evergreen upgrade and says, "Wow, that was great. I want to go back to the old crappy way of doing things." Nobody does that, right?

And so when we think about the stickiness, when we think about the customer retention, that Evergreen affords us is amazing. If you think about this just from a purely transactional point of view, we don't have refresh cycles, right? We -- I love refresh cycles. A competitive refresh cycle is a big opportunity for us. It's a sales opportunity, right? With Evergreen, we don't afford our competition, the same favor. Now you couple that stickiness with network effect. You couple that stickiness with what we've done with the intelligent control plan with Fusion. We've now introduced a set of capabilities such that having 1 array is great, having 5 arrays is great. we're actually now at a point in time where as you have the fifth array as you have the 50th array, buying that 51st array is actually incrementally better than it was to buy the first one. Because it's easier to manage, it adds to your pool of resources. It is -- it just fits very, very nicely into that cloud operating model.

Number three, we're continuing to expand our technology differentiation across the board. I'm going to go deeper into a couple of the areas. A couple of these areas in the next couple of slides. Charlie talked about what sets us apart as a unique sustainable advantage is that we are a high-technology company. This drives differentiation. And we do this intelligently, right? We're now operating in a size and a scale across multiple markets that we're able to take those technology investments. The outputs of those technology investments and leverage and use those in multiple different ways, which affords us a tremendous amount of leverage and scale.

All right. So we spend a lot of money in R&D. One of the things that I want to highlight here is most of our R&D spend, most of our resources and our efforts are actually directed towards software, right? So well over 95% of our R&D spend actually goes to software. And that's quite intentional. It's intentional because most of the value, most of the differentiation we deliver is actually software driven, but it also means that because it's software-driven, we can go and deliver that on multiple hardware platforms across the entire portfolio. That results in an expanding set of differentiation. And we only have a certain amount of time. I'm not going to go through the entire list. Charlie flashed some industry accolades. You look at the Gartner results, the Forrester results. These are all third-party validation of that differentiation that keeps growing in time, right?

Usually, you see these things clump towards the middle. You see us moving further and further apart. We do this in a number of areas. Flash management, DirectFlash. We'll talk about that with the hyperscale solutions is one of the key areas. The other thing I'd call out, though, is that no matter what we're talking about, whether it's filling out the portfolio, whether it's particular features, we're not chasing competitors. We're not just trying to build a better version of a -- better version of somebody else's feature. We're trying to solve those problems better. And we can point to numerous examples, whether it's in terms of data protection, what we've introduced with SafeMode. We can look at storage management, what we've introduced to Fusion. We're investing heavily to find better ways for customers to do things, and we're getting rewarded for it.

We're getting rewarded for it because customers are seeing the outcomes. So customers are seeing the outcomes in increased efficiency, better reliability, overall significant TCO savings, massive, right? And this is a virtuous cycle. We invest in R&D, that grows differentiation, we deliver better outcomes. We drive growth, we reinvest in R&D, simple story. One more thing to point out here. We have a secret weapon, which is Evergreen. With Evergreen and Evergreen NDU, we make it a lot easier and more appealing for customers to take software upgrades. We can accelerate the cycle in ways that nobody else can. Think about it this way. Your typical enterprise vendor builds a feature, they ship it, customer says, okay, that's great. I'm going to wait for the second major version until all the bugs and things sort of worked out 2 years past, they finally deploy it. Think about that time to value realization from the R&D investment to when the customer gets value to the rebuy, it's a really long cycle. We shrink that thing.

All right. So we're gaining share. market is also growing. I mentioned before, I would say 2 major effects that ultimately combine to accelerate the growth of the enterprise storage market. Historically, this market has grown, call it, mid high single digits. We see the market is accelerating growth over the next couple of years to about 12% CAGR. I'd pull out 2 drivers for this, both relating to AI. One, AI is just making data more valuable, right? We produce a ton of data as an industry. Most of it today doesn't get stored because it's not valuable enough to store. Customers aren't getting results from it. as AI becomes more prevalent, data becomes more valuable, it's got to be accessible. So we see that as constructive for the market growth.

But number two, new use cases. As I becomes or as enterprises focus more on data sovereignty and control and bringing AI in-house. These new GPU connected arrays, what we're calling Core AI use cases are on the rise and are going to add to the historical -- what we've historically looked at as the enterprise storage market.

As an example of this, I'll highlight one of the examples that we have seen with a major multinational bank. And this has been an effort that started in their labs, has gone through pilot initial stages of production and is now going to full scale across the entire bank. This is a large bank that effectively has said, "Hey, for all the work that other people do going to Claude or OpenAI to build agents or tools, we don't have any of that. We want to do that all in-house. We want to build a system for our employees, the bank's employees to build agents, tools, the whole 9 yards, and we want to run it all in-house.

They chose Pure to back this at multiple levels, both the data going in, the agent memory, the user memory, the KV cache as well as the agent development environments. Now you might ask what was it that drove them to choose Pure. Well, one, we were deployed as part of an NVIDIA reference stack. And so they went to NVIDIA. The looked at the entire ecosystem. But what made us stand out apart from the pack is that -- and they had looked at competitors. What made us stand out from the pack is that we could deliver not just the performance well above the performance that they anticipate meeting, but we could do that in a way that no other vendor could coupled with all of the enterprise capabilities because these are mission-critical environments, right? Security, reliability, availability, all the things that we're known for, nobody else on the market could bring both of those things together.

All right. So if we step back for a minute and we look at the core business, core plus core AI, growth equation is fairly simple. We're going to keep gaining share, going to keep reinvesting to gain share and accelerate the pace of that. The market is growing. Those 2 effects compound to drive significant growth in our core markets.

All right. So let's talk about some new stuff. As we look beyond the core, 3 distinct new high-growth market opportunities in areas that we're turning our focus increasingly to. Number one, modern data software. We'll talk about scale AI. And then last but not least, I'm sure we'll go deep into hyperscale solutions as well. With modern data software, I would say this is -- in many ways, we're in the early stages of this journey, right? In fact, we were -- our first foray on this path, I would say, was actually Portworx, right? With Portworx our thesis was and still is that applications application environments are fundamentally going to change. They're going to change over time, and we want to be on the forefront of driving that change. When we acquired Portworx 5 years ago, 5.5 years ago, that change was largely driven by containerization, open source, Kubernetes all of the -- all of which are still true.

You now have other secular forces coming to play, coming together to accelerate what we believe will be a 5-, 10-, 15-year journey, significantly up ending the application environment in the enterprise. I talked about this a little bit in the morning, but AI plays a big part, right? AI forces a rethink about -- rethink of how traditional enterprise architectures are built, how software is built. The focus shifts to data, the focus shifts to making sure that AI and AI-based systems are getting fed the right data, that it's organized and governed properly. These applications are built entirely differently than monolithic traditional enterprise applications. They're built on containers, Kubernetes, making use of open source. And also, there's an importance now, right, as we talked about this morning, our goal, where we see the future heading is being able to work with -- a customer being able to work with their data, no matter where it sits. Data sits all over the place. It sits on-prem, sits across multiple prem environments, sits in the cloud, across clouds. And so we need to be able to help serve customers where that data sits across these different environments.

So, if we unpack this, and Prakash talked a bit about this morning and Charlie hit on this as well. The first leg of this is meeting the needs of helping customers identify what the right data is. Most of the time, when you go talk to a customer, when I go talk to a client in the enterprise, they'll say, "Hey, I want to do all this great stuff with AI. They'll say, "Hey, that sounds awesome. Where are all of your data sources today?" And 9 times out of 10, the customer will probably chuckle at me and say, I have no idea. 16 different places. Every department has their own database. We have multiple copies. So problem #1 is figuring out what is it you have, where does it sit? And what does it mean? This is where One Touch comes in. This is where data intelligence, where we're headed with that. And ultimately, it's about helping customers in a very automated way, AI-assisted, find catalog and understand what the right data is, what it means and understand how best to use -- what are the right ways to use it and feed it into AI systems. Garbage in, garbage out. You got to feed the right data in, we're going to go solve that.

Number two, once you find and understand what the right data sets are, well, it behooves you to organize it in a way to make it easy to work with, easy, efficient, fast, so with data stream built on NVIDIA's AI data platform, we've brought all of the pieces of data organization for AI together, automatic ingestion, vectorization and in organizing and providing it in a way to RAG and inference systems that both optimizes the performance as well as preserves all of the data security needed across the board from the source data all the way to the vectors.

All right. So we have a couple of capabilities. We're going to deliver the software above and beyond and outside of our arrays across multiple environments. But if we look at the broader application stack, application stack looks very different than traditional enterprise apps. They're built on containers, built on Kubernetes. And at the same time, a lot of the traditional apps are moving to containers. And so with Portworx, we're now seeing increased demand for Portworx, both in terms of virtualization modernization as well as supporting data pipelines, data analysis and inference systems. And then last but not least, there is a greater desire and a greater need to be able to connect data sources and work across clouds, across prem and cloud. And so this is where Everpure Cloud, the evolution of Cloud Block store into a managed service and Portworx come together to be able to provide that multi-cloud capability across both traditional applications as well as modern applications.

All right. So shifting gears, we'll chat a bit more about the scale AI opportunity. This is a focus now on the neoclouds, the largest AI model builders, the Frontier builders and the tech titans. And simply put, this is where the needs of AI environments we see in the enterprise are well served with our existing solutions. If you scale that up by 10 or 100x and you look at the neoclouds, that's a whole other level of scale and performance and demands that traditional solutions are not well placed to serve. At the same time, neoclouds are clouds. They have reliability constraints. They have SLAs to meet. So the balance of getting the right performance levels and the right manageability and reliability becomes of utmost importance, and we're the only ones that can go provide the balance of those 2.

The last point I'll make here is with the scale offerings led by FlashBlade//EXA and Charlie hit this a little bit upfront. Coupled with FlashBlade//S, we're now the only vendor that is able to cover the entire spectrum of AI needs from the smallest labs to the mid- to large enterprise all the way to the neoclouds. And then in my next section, we'll talk a little bit about the hyperscale.

So let's put some numbers on this. We talk about core AI. We talk about what's happening in the enterprise. What we typically see in the enterprise in the example I gave before is an enterprise might deploy an environment serving tens to hundreds of GPUs, maybe a petabyte to 10s of petabytes of data, supporting hundreds, maybe up to 1,000 users or so. FlashBlade uniquely fits the spot well. We provide that entire range of performance scale capacity and the flexibility to grow and move within that space to serve RAG or inference environments, agent deployments. And we do this with an all-in-one package, including the storage, the software as well as the networking.

As you look at the neoclouds and you look at the concentration of demands that are happening in the neoclouds, think about scaling this up by 2 orders of magnitude, 10x, 100x, right? You're not serving tens of hundreds of GPUs. You're serving thousands to hundreds of thousands of GPUs. You're serving hundreds of petabytes to exabytes of data. You're serving tens of thousands, maybe hundreds of thousands of users. Now you might say these are completely opposite ends of the spectrum, right? They're completely different demands. What they're not is completely different products, right? What we've been able to do with FlashBlade//EXA is extend the same technology we use to serve the enterprise with FlashBlade//S, the software that drives that pair that with a more open hardware platform to meet the needs of this new level of scale. So complete opposite ends of the spectrum, same technology that allows us to address both.

What does that look like? Well, Charlie introduced this a little bit upfront. Really, 3 things that we're leveraging, using extending from the core FlashBlade and Purity software. Number 1 is the utmost performance we can deliver; two is, the scale and flexibility; and three, is that mission-critical bulletproof reliability. On the performance front, -- everybody tends to focus on bandwidth and how many bits per second gigabytes per second terabytes a second. And I apologize for those in the room who are with us in the morning session, but I'll draw an other analogy. It turns out in AI workloads, both data performance and metadata performance are important.

What do I mean by that? Well, think of it this way. If I ask you to -- if you ask a computer to open a file, the computer has to do some work to figure out where the data sits and then it has to do the reading. The reading part is the data performance, the finding the file is the metadata performance. Okay, it doesn't sound too bad. Let me draw a more human analogy. If I handed you a book and go to a page and ask you to read the page, it might take you a minute or 2, and not too bad. I hand it to a speed reader, might take them 20, 30 seconds. Wow, that's a whole heck of a lot faster. If, on the other hand, I'd point you to a shelf of books and I say, "Hey, go pick the third book off the first shelf, turn to Page 17, read 2 sentences, pick another book off the fourth shelf, go to random page, read a couple of sentences. The more work I ask you to do to find the data versus reading the data, right? The more metadata work I'm asking you to do versus data work.

And so what happens in AI workloads is you need a balance of both these things, right? If I ask you to go pick a whole bunch of random books and read a sentence or 2 from each we can agree that it doesn't really matter how fast you read, you're going to spend all your time finding where that data sits. We do a really good job of both of those things. We do a really good job of both of those things because of the software. We've built FlashBlade. And with FlashBlade//EXA, we've been able to extend that and decouple that to create independent scale of the ability to supercharge the performance for both legs of both metadata and data.

We've been able to also extend the scale in terms of capacity, right? So really now with FlashBlade//S and EXA covering the entire gamut of tens of terabytes to tens of exabytes tens of gigabytes a second to tens of terabytes a second and everything in between. And then most importantly, and I mentioned this before, as we think about neoclouds, GPU utilization, reliability, usability, operational simplicity are utmost concern. We've been able to do this on open standards without a whole bunch of tinkering without a whole bunch of nerd knobs for customers. We've been able to do this so that neoclouds is going to operate these environments on thin staff in a way that gives them mission-critical reliability that we're known for.

Some results. So MLPerf is the industry's leading benchmark when it comes to AI systems, in particular, storage. These are results we recently published with the MLPerf, the latest MLPerf 3.0 benchmark. These are standardized tests. What our submission has shown is that we, number one, across all categories and all configurations. Somebody asked me over lunch, "Hey, your peak throughput for checkpoints went up last year. You did 10 nodes, you now do 30 nodes, it's up 3x. Why is it only 3x?" I said, "Well, it's entirely linear, right? It's entirely bounded by how much hardware we decided to buy. If we were to deploy twice as many nodes, that number would go up 2x. And so this is exactly the power of taking the software scalability and the underlying platform know-how that we have with FlashBlade//S and taking it to the next level of scale. And giving our customers the neoclouds the same predictability that the enterprise has had for the last 15 years.

All right. So if I zoom back out and look at the neoclouds, a couple of things. One is these large-scale AI environments. 3, 4 years ago, the focus was entirely on performance. Now yes, they need the performance, but 2 things are happening. They need performance across an increasingly varied set of workloads. It's not just large model training. It is training. It's also inference, it's also agent deployment. It's all of those things happening at multiple -- at the same time, it's all of those things happening at the same time across multiple tenants. And so they need all of those things with the same reliability, same tenant isolation, same security all the enterprise capabilities start creeping into these environments. And this is an area where we're uniquely positioned to be able to meet both ends of those needs in a way that nobody else in the market is able to do.

All right. All right. So what everyone's been waiting for. Let's talk a little bit about hyperscale. So as we look at the hyperscale solutions, 3 takeaway points here. Number one, we deliver significant technical and structural advantages to the hyperscalers. Two, it's not just the point technology. We deliver them a compelling solution that gives them a consistent, unified single architecture that gives them a consistency across performance tiers as well as flexibility to adapt to new workloads. And then number three, there's a race to deploy flash and we present the most compelling option to get their fastest.

So why is that? Well, if we look at what's been happening over the last 5, 10 years in hyperscale storage, and I simplify this to big animal picture is a couple of things. One, performance demands keep going up, thing called AI. Two, hard disk drives, relatively speaking, keep getting a lower. Yes, they might get faster -- sorry, yes, they might get larger, but they're not getting any faster. So performance per terabyte is actually going down, that gap to flash increases. And so natural to see that there's been a rush, there is an increasing rush to deploy QLC flash, right? Fairly simple equation.

So what does this result in? Well, it means that every hyperscaler out there is doing a couple of things. One, they're racing to secure SSD supply QLC in particular, as fast as they can. They're doing this across multiple vendors, which then puts the onus on them to do multiple qualification cycles with different SSDs, for multiple vendors, and they're doing this at a point in time when their workloads are changing rapidly. And then number three, they got to make this whole thing efficient, right? They're at the same time trying to figure out how to get power, how to get space, how to build new data centers, they've got to drive efficiency out of these investments at the same time. So there's a lot going on.

So what are their choices? How do they get there? Well, there's a couple of pads, right? Option 1, Option A is what's called the status quo, right? They can continue to build out based on conventional SSDs. This is what they do today. They're going to keep doing some of this, but it's getting increasingly inefficient, right? The SSD architecture has inherent limitations, as they deploy more, as the workloads get more performance demanding and more varied, those limitations get stretched even further. As they have to qualify multiple SSDs. There's more work on the front end. So this thing is getting harder, but they're going to keep doing some of that.

Well, they could try to change the approach instead of working with conventional SSDs, they could try to build substantially similar technology to work directly with NAND, and they tried to build that in-house. Could they do it? Yes. Will they do it? I don't think so. Why? Well, it's a lot of R&D. We've been at this for 15 years. We have some of the most sophisticated knowledge in the industry even including the NAND manufacturers themselves and how flash works. And we have an enterprise business that pays for it, right?

When you think about the R&D to develop the technology, when you think about the ongoing maintenance and qualification of each new NAND part from a vendor, much less each new vendor that gets onboarded, that's a lot of R&D to get started. It's a lot of R&D to continue investing. You can do it, but you have to consider opportunity cost, right?

They have option 3. Get the boost of both worlds. They can get the technology benefits, working directly with NAND, don't have to take on the onerous R&D, work with Pure who now has a proven technology, integrated multiple hyperscalers. And this is really how we see the opportunity playing out.

All right. We've talked about this in the past a little bit of a rehash, not going to go deep into hard drives. They're terrible. They're still terrible. They're getting worse. SSDs versus DirectFlash. So at the end of the day, the structural inefficiencies that are inherent in SSDs come from the fact that SSDs are effectively a technology coping mechanism. They're coping mechanism to expose semiconductor NAND flash to software, computer software in a way that makes it look like a hard disk drive, right? In order to do that, you actually have to do a bunch of work inside the SSD. There's complex firmware. There's circuitry to make that firmware work. There's a little controller chip, there's DRAM to make the whole thing. It's basically a little computer in a box. That thing keeps getting more and more complex, every generation of NAND as the memory manufacturers make denser NANDs, stack more dies, it gets harder. That thing is bursting at the seams.

Well, what does that mean? It means compromises and reliability, means compromises in efficiency, it means compromises in performance and performance predictability. We bypass all of that. We bypass all of that with our software, which we run at the host level essentially treating flash the way it's meant to be treated as a semiconductor. And that's what we're delivering with DirectFlash. But it's much more than just the technology. It's much more than just direct flash. It's how we've packaged the solution to fit within the hyperscaler architecture. So I'm going to go deep into this a little bit. We've talked on the earnings calls a little bit before about how the hyperscalers design storage in a horizontal tiered fashion.

And we now have conversations with multiple hyperscalers that all substantially look very similar to this. At some layer, they have their own distributed storage software that runs across multiple storage nodes. And underneath this storage software layer, they design specific hardware storage node combinations, depending on price, performance, capacity needs. They typically have multiple tiers, hot tier, high performance, a warm tier, pretty good performance and a cold and archived tier, best economics possible. Typically, you'll find SSDs at the top hard disk drives at the bottom and then maybe a combination of both in between.

This is kind of the starting point. This is how hyperscalers have traditionally designed their storage environments. What we've done is 2 things. One is we packaged our DirectFlash technology, which I just explained, has all these benefits over SSDs, we package that technology in a way that fits very cleanly, integrates very easily into their distributed storage software. It's not a big -- it's not hardly any software change at all on their distributed store software. We slot in just like their existing storage. We just do it much better. We do it much better because of the attributes I just described. We also do it much better because we give them a consistent architecture to be able to serve the different tiers, whether it's hot, warm or cold with 1 architecture dialed up and down based on software configuration and media configuration, right?

When you think about that, economies of scale, right? Chad talked this morning about how the enterprise benefits from operating in the same way across multiple environments, huge benefits. Think about what that means at the hyperscale, massive. But wait, there's more. The other thing that's happening is it's not just a set of clean tiers anymore. With AI, with the advance of AI-driven workloads, even the hottest tiers are now getting stressed for performance. As new workloads appear, new applications appear above the hyperscaler storage software, pushing performance bounders beyond what a single SSD tier can provide. So the hyperscalers are now being forced, they're being pushed in a direction of designing multiple different hardware configurations for each of these, what used to be 1 tier. They can do that. But again, it's more work. It's more inefficiency, it's stranded capacity, it's one more thing to operate.

And here's again where our single consistent architecture can now give them flexibility, software tuned, media, configured but entirely consistent architecture and flexibility to meet not just all the performance tiers, but adapt to new workloads very, very quickly. And so ultimately, the 2 value propositions we're delivering to the hyperscalers is the core technology which gives them all these benefits, performance, reliability, efficiency but then the packaging, how we're delivering it in a way that gives them operational consistency across multiple environments, whether it's SSDs now in time as prices drop all the way down to disk drives, as well as adapting to the needs of new workloads.

It's not just the technology. There's a lot behind this. There's a lot of history. There's also a lot of vendor relationships and well, how about flash. We've been at this a long time. We work with multiple NAND providers, key partners that -- like KIOXIA that we've worked with over a decade to help co-engineer and co-design future road maps. We were with multiple NAND generations across multiple suppliers. We do the qualification. We do the reliability engineering. We've got an enterprise business to go pay for that. We can go and offer the benefits of that hyperscalers.

I mentioned qualitatively some benefits, not going to unpack this slide, but just to put some numbers on it, right? When you think about twice the lifetime, we think about 5x reduction in footprint, when you think about 2 to 5x improvement in reliability, these are serious numbers. These are serious numbers no matter who you are. These are serious numbers to an enterprise. Think about operating at hyperscale, right? The operational savings, the headcount reduction, the reduction in tickets, the improvements in SLAs, these are huge benefits.

And so at the end of the day, right, if we simplify this, the 3 kind of things that hyperscalers are looking to us, looking to our solutions to provide, number 1 is the efficiency, whether it's, again, equipment efficiency, utilization, power, reliability, it's the uniformity. It's how we've packaged that solution, that technology to give them 1 approach to deal with multiple tiers, multiple uses, multiple needs. And it's our supplier relationships and diversity. Our ability to give them access to the bulk majority of the world's NAND supply without having to do and bifurcate their qualification work.

And it's because of these things that have driven our first couple of top hyperscaler wins as well as now an increasing set of interest beyond the top 5 hyperscalers for this set of solutions.

All right. So I'm about a time. I'm going to try to wrap this thing up. I've hit a lot. But I think the takeaway message here is simple. We're now at a point of inflection in the company where we're pursuing not 1, not 2, but 4 different high-growth market areas that are expanding into new areas of software, extensions of our hardware technology and everything in between and it's taking us into significantly new and larger market segments. And we're doing this not as a bifurcated set of efforts. We're doing this based on 1 foundation, same technology, 15,000 customers, validation at the largest scales driven by 1 product organization, 1 architectural foundation set of IP and 1 sales organization. Thank you.

All right. So now for the session you all have been waiting for. I have the pleasure of welcoming my friend, Tarek to the stage, who will take you through how all of this comes together to influence our views on our longer-term financial model.

Thank you, Rob. You ran the clock like a true football coach, amazing. Thank you very much. Market is closed. So now I can freely speak, and I'm delighted to be here with all of you. So -- so good afternoon. and welcome, everyone. For those who don't know me, I'm Tarek Robbiati, I'm the CFO of Everpure. And I've just been in the seat for about a year. And I have to say I'm having a wonderful time. Someone asked me at the break, how do I feel, given my past career, I reiterate, I'm having a wonderful time. And most importantly, I remain deeply impressed by our massive market opportunity. And the team executing against it. As Charlie noted, we treat data storage as an essential technology enabling customers to manage their most critical assets in this day and age, and that is data.

Supported by a collaborative culture driving infrastructure modernization we are literally redefining the industry. And I think you are seeing that come through today. Today, I will outline the foundation we've built for durable growth, detailing our long-term financial framework for growth, profitability and capital allocation. We will begin with our current business performance. Then I'll discuss how strategic investments are expanding our addressable market to deliver sustainable, profitable growth and long-term shareholder value.

You heard it from Charlie, we are at a major inflection point. Our business is accelerating. It is built on a strong multiyear foundation of differentiated software, a recurring customer base and a culture of innovation. Today, multiple opportunities are converging to drive a new growth phase for our company. We are redefining data infrastructure and expanding into high-growth markets like scale AI, modern data software and hyperscale solutions. What's very important to me and crucial to me as CFO is I see us pursuing these opportunities while maintaining very strict focus on profitability and financial discipline.

What I'm going to tell you is going to be broken down in 3 chapters I'd like to cover today. First, we'll deep dive into the current business foundation. Then we will evaluate our ability to expand our core and capture share in new high-growth adjacent markets. I'll tell you the potential of each one of those markets. And finally, and most importantly, I will share with you how our financial profile is resetting and resetting durably. This is not an incremental change. It's a reset, and we believe this reset is durable.

So I've got to start where we finished last time we met. That is our Q2 results. Our Q2 results, any way you want to look at it was nothing short than outstanding. We grew revenue 38% and operating income, 77% year-on-year, beating guidance and consensus. For FY '27, and as you already know from the guidance we provided and reiterate today, we target at the midpoint, $5.05 billion of revenue is implying a 38% year-on-year growth and $950 million in operating profit that is implying a 50% year-over-year growth.

According to Jefferies, Everpure is 1 in 7 companies in the tech software space expected to grow revenue more than 30% in the next 12 months. Everpure is firmly in that club. Plus, we have the operating leverage enhancing our model to drive even faster growth in our bottom line. And no, this is not a flash in the pan, okay? Just to make sure you're clear. You know, you probably heard that Everpure addition earlier this week into the S&P 500 Index is a testament to delivering strong growth with margin expansion consistently quarter after quarter over many years. This is what the profile of a high-performing company is.

And that brings us back to the inflection point. This top line acceleration reflects strong demand, deep customer engagement as they rethink their infrastructure requirements. So you know this already, but it's important to put some perspectives. So let's frame our performance in a longer-term context showing our revenue trajectory from fiscal year '22 to fiscal year '27. Over this 5-year period, we expect an 18% revenue CAGR that is really premium growth in a historically cyclical market.

Looking to the right of the chart, the expected growth for fiscal year '27 accelerates to 38% year-on-year. There are 2 key takeaways. First, this is not a new growth story. You heard it from Charlie, we delivered durable growth for years. Second, and this is news maybe, our growth trajectory is now accelerating. This inflection is the result of a very strong foundation and continuous investment in our software platform, our technology, customer relationships and product velocity. We believe this faster growth will continue across economic cycles because it is driven by structural tailwinds in how customers build and manage data infrastructure which we will address shortly.

Also when you look at growth, not every growth is created equal. Quality growth is really important. Cyclical growth is not quality growth. Examining our performance in detail reveals increasing revenue quality, driven by our SaaS offerings, particularly Evergreen//One. That's the market's only true SLA-based storage-as-a-service solution. Unlike traditional product sales, which can be more directly affected by component cost fluctuations, Evergreen//One is built on long-term customer commitments with lower upfront capital requirements. With Evergreen//One, customers can ramp into growth and are built on a consumption basis, which allows them to better match expense outlays to the growth of their solutions.

Two points worth noting. First, Evergreen//One has gained tremendous traction in the current environment as we are able to contain prices. Two, and because we control the configurations of the solutions that underpin the Evergreen//One SLA-based contracts. We are able to fully manage the margin of our Evergreen//One offering. The key takeaway is that as of the second quarter of fiscal year 2017, Evergreen//One's TCV is on track to surpass an annual run rate of $1 billion. And that is driving sustained revenue growth and earnings quality for the foreseeable future.

As you know, if we look at the bottom line. And when you look at the bottom line, you will probably realize that we prioritize long-term profitable growth rather than growth at any cost. From fiscal year '22 to fiscal year '27, operating income is projected to grow at a 32% CAGR that is outpacing our 18% revenue CAGR we spoke about a moment ago. For the current fiscal year, operating income is expected to increase by about 50% to $950 million at the guidance midpoint, demonstrating that we are growing operating income significantly faster than revenue and therefore, creating operating leverage.

As we scale, we will continue leveraging past platform investments and allocating R&D towards high ROI growth opportunities. This is the crux of the business model that Charlie spoke about, reinvesting to accelerate growth remains central to our financial philosophy. And in so doing, we ensure that those investments translate into durable revenue growth and increasing profitability. Ultimately, our long-term builds a virtuous cycle where growth and profitability reinforce each other with profits fueling the continued enrichment of the Everpure software platform.

I want to hammer this point or the third time. This is an R&D-driven model. Hopefully, you will realize that, but it's very, very important. It brings me to the most important differentiation factor of Everpure, and that is innovation. You heard it from Charlie and Rob already. On a non-GAAP basis, we invest an industry-leading 19% of revenue into R&D annually. With all engineering talent focused on a single platform, that's key if you have to spread your dollars over multiple platforms, you're wasting time and resources and dollars.

Here, all the engineering talent is focused on a single platform for faster time to market and better ROI. Our software-centric approach backed by skilled hardware engineers, designing simplified products, deliver superior software intelligence that extracts maximum performance efficiency and value from the underlying NAND. We've used this approach to enhance the customer experience across many areas. You heard that. We pushed innovation in data management, in power and space efficiency, automation and also enabling new business models like Evergreen//One. The key point to take away is that our R&D investments fuel future growth creating product velocity that penetrates markets and drives durable revenue and profits for years to come.

So let me summarize what we discussed so far in the first chapter. First, our strong core business has delivered durable multiyear growth, which is accelerating to 38% revenue growth in fiscal year '27 compared to an 18% CAGR since fiscal year '22. Second, we are driving significant operating leverage with operating income projected to grow about 50% this year, and at a 32% CAGR from fiscal year '22 to fiscal year '27. Third, our continuous innovation strengthens our modern tech stack to better serve our customers and sustain revenue growth and profits. In short, the Everpure value proposition combines a differentiated technology platform, durable core growth and resources to fuel acceleration.

Moving forward, we plan to capture share in high-growth adjacent markets with large untapped potential translating our R&D leadership into long-term financial performance. So before we talk about the expansion of our addressable market, I want to spend some time on a very important topic that I'm sure is top of mind. It's a very important dynamic that is impacting our industry right now, and this is related to the significant increase to component costs. This chart illustrates the magnitude of the NAND and system cost increases since the past 12 months from the third quarter of fiscal year '26 to the third quarter of fiscal year '27. The increases we are seeing across both QLC and TLC NAND as well as overall blended system COGS are truly unprecedented. While price adjustments to offset these costs have aided reported revenue growth, we are not and never have been a price-driven growth story. We have not and never have been a price-driven growth story.

Our momentum is primarily fueled by strong underlying demand, market share gains and expanding customer footprints. So as you look at our growth trajectory, it's important to distinguish between the acute benefit of pricing and these far more sustainable growth drivers.

And the next slide provides some historical perspective on that. You've seen this chart already in Charlie's presentation, this fantastic line with the uptick, as Rob noted. We've added a new lens for you to identify the periods where NAND costs were inflationary and deflationary. The vertical bands on the chart identify each period. And as you can see, putting today's environment into the historical context, we've demonstrated since 2013 that we reliably gain share through every NAND cycle, regardless of whether input costs are inflationary or deflationary.

And here's why. In inflationary environment, as total investment costs rise due to trends like AI, what do customers do? They prioritize quality and upgradable products to protect their investment ROI. Conversely, in deflationary environment, our differentiated architecture leverages falling component costs to drive more efficiency, while customers also optimize for inflationary TCO factors such as labor, power and space. Ultimately, we do not view NAND pricing as a determinant to our growth. Let me repeat that. Ultimately, we do not view NAND pricing as a determinant of our growth. Our gains in share are driven by the enduring value we provide to customers. With this growth proven durable across multiple cycles, you may wonder what happens to your gross margin. So let's getting to that.

The simple punch line here with the same analysis at the gross margin level is that we are gaining market share without compromising our economics. We're maintaining a substantial product margin lead over our nearest public competitor across all market cycles. We control our gross margins. You hear me? We control our gross margin. And we do so through disciplined, deliberate pricing and balance near-term performance with long-term growth by delivering customer value while adapting to fluctuating cost structures.

Ultimately, our financial model creates substantial operating leverage capacity, one that builds a durable high-quality business that grows share, leads in margins and increases profitability at scale. You may say, why is that? So why do customers choose Everpure I want to bring you back to the beginning. Our differentiation starts with software. And as Paul always says, it's all about the software. Look, there's nothing special about having access to the underlying NAND itself. SSDs do that all the time. The secret sauce is providing customers with the best mix of software and architecture to extract the most value out of that net.

As Rob and Chad noted earlier in the morning, Everpure cut power and space by up to 5 times delivers over 10x greater reliability requires up to 10x less labor features constant upgrade with our Everpure//Forever program. And that is driving a 50% lower total cost of ownership or TCO.

So now think about the sheer economic value that this delivers for customers running large data centers. Those customers buying decisions are ultimately based on the total cost of operating their infrastructure over its useful life. Most customers are not asking who has the lowest price per terabyte. Most of them ask what is the most efficient way to run my data center infrastructure. And by drastically reducing power footprint, labor and complexity, our solution delivers compelling long-term economics despite a higher initial purchase price.

And here's what you can conclude. In other words, if you look at our gross margins, our customers reward us with a premium for our solutions because we lower their total cost of ownership. And this drives industry-leading NPS durable market share gains and margin leadership.

So moving forward, this is news. We will track our progress in the core and new markets that were identified by Charlie and Rob. We have a solid foundation with our core and have multiple paths for future growth. While continuing to report revenue and subscription revenue as we do, we are organizing our revenue into 2 main categories to help investors track our progress with respect to the expansion of our business model.

The first category Core and Core AI encompasses established platforms that you know already, FlashArray, FlashBlade and Evergreen//One, supporting both CPUs and GPUs representing our ongoing growth and market share opportunities in the core. The second category, which we're going to call new revenue, introduces 3 growth vectors. Scale AI, you heard from Rob, it includes our excess solutions for neoclouds, AI Titans, et cetera. Modern data software, which includes Portworx, Pure Protect, Everpure Cloud, Data Intelligence, also known as OneTouch and Data Stream and hyperscale solutions, which leverages our DirectFlash technology for major hyperscale customers.

I would like to reiterate that these are interrelated businesses that are natural extensions of our R&D capabilities. I know you hate changing your models. You're not going to change your models. This is the good news.

Moving forward, we'll keep the same reporting segment. But in addition, we intend to provide you at the end of each year with annual updates about the growth in new revenue, which will track the expansion of our business beyond the core. So how do we do new revenue within our overall market opportunity. Historically, every publication around the storage market was referring to a $50 billion market, growing about 9% per year. However, AI and data proliferations have dramatically expanded our addressable market, offering the age of data primacy driven by 2 structural tailwinds. With exponential data growth and massive AI-related data infrastructure deployment. So to put things in perspective, in 2025, the world generated 173 zettabytes of data. I don't know how many of you know what a zettabyte is, it doesn't matter, but it's a 1 followed by many, many, many, many, many zeros. Anyway, it's a 90-fold increase since 2010, where in the world was generating only 2 zettabytes of data. And this has been driven by IoT, AI workloads, video surveillance, cloud adoption, real-time processing, you name it.

And in parallel, concurrently, you know this, you research the market better than we do. Cumulative CapEx for hyperscalers related to AI infrastructure is estimated at about $2 trillion to $3 trillion over the next 3 years. that is 1.5x to 2x the U.S. defense budget. So this helps paint the picture as to why we're so bullish about Everpure growth opportunities in the core.

Now let's examine also the projected addressable market sizes Everpure capture over the next 3 years beyond the core. So we said already that the core is accelerating at a 12% annual growth rate, reaching $79 billion by fiscal year 2030. Here, we compete with traditional enterprise storage providers like Dell, NetApp, HP and IBM. In hyperscale solutions, which I'm showing you on the chart, based on our estimates, we estimate that TAM will grow at a 22% CAGR to $97 billion from what it is today at $44 billion, $97 billion by 2030. And Here, we are competing against internal cloud solutions, SSDs and hard disk drive providers. Our modern data software TAM, spanning app modernization and data intelligence is projected to grow at a 26% CAGR to $22 billion by fiscal year 2030.

And here, we face different types of competitors. NetApp, Dell, Veeam, Cohesity, IBM and specialized providers like Databricks. And then finally, Scale AI is projected to nearly double to $9 billion by 2030, competing against traditional providers, high-performance storage and AI specialists such as VAST, DDN and WEKA. So we compete in 4 high-growth markets against different players in the industry. And these estimates, I'd like to underscore this is really important. It's very important to understand that these estimates are based on our current solution set with potential for further TAM growth as we expand our current solution set and customer adoption and uses continue to increase.

So let me summarize the key points to remember here okay? Our core and core AI business opportunity is now growing at an accelerated double-digit as a result of the explosive growth in data and AI deployments. Number two, Everpure modern tech stack and rapid product velocity has us very well positioned to compete in several new faster-growing markets, hyperscale solutions, modern data software and Scale AI. And our total addressable market opportunity over the next 4 years is increasing by a whopping almost $100 billion from $108 billion to $207 billion.

Now of course, the TAM is only useful if we have a credible right to participate in that market. That's why I also want to show you our serviceable addressable market, or SAM. Our serviceable addressable market, SAM, focused strictly on all-flash technology because that's all we do is doubling from $34 billion in '26 to $69 billion in 2030. And that is driven by moderate commodity pricing at some point in the next 4 years and AI-created workloads. Anchored in our all-flash DNA, differentiated technology platform and deep customer relationships. Our strategy extends our core advantages into faster-growing data infrastructure and data management markets.

With the strength, our ambition is clear. and is to be the #1 provider of all-flash solutions uniquely positioning Everpure to compete and win across these 4 markets. All right. So I think I need to relax a little bit because now we're getting into the meat of the presentation. I'm not sure, Paul, my ears are ringing. I'm hearing everybody saying Jerry Maguire moment, "show me the money." All right? That's what I'm hearing you say. All right. So I'm going to show you the money.

We spent the last several slides explaining why we believe the opportunity is expanding. Now I want to give you our preliminary outlook view for fiscal year '28, okay? The quick punch line is that Everpure is entering fiscal year '28 with significant momentum and an expanding set of growth opportunities. Building on our fiscal year '27 revenue outlook of 37% to 38% growth per year, our preliminary fiscal year '28 outlook is here. Is $7 billion to $7.3 billion, implying 39% to 45% growth year-on-year. That's right. Someone said, wow. This indicates accelerated top line growth, delivering a 2-year CAGR of approximately 38% to 41% and for fiscal year '28 -- to fiscal year '28. This is well above historical levels and current consensus.

I want to make sure you're level set on this. This premium growth is driven by the diversified portfolio we spoke about with Charlie and Rob continued core strength, emerging hyperscale solutions, expanding into modern data software and scale AI. And then here, I'm asking you to cut me a little bit of slack. Because it's unusual for me to provide you guidance for the next year at the middle of the current year. So this is not guidance, but it's preliminary outlook. And its share to reflect with you our strong visibility and confidence in this trajectory. It is an outlook. We stand by that. but we will provide further details about it and formal guidance consistent with our usual timing as we wrap up fiscal year '27 and enter fiscal year '28. The range is wide and that is because we are in the middle of our planning cycle. And when we complete our planning cycle, we will come back to you with a narrow range. That makes sense?

Okay. So now the question everybody wants to talk about. What about your operating margin? All right. But good news is not limited to accelerating our revenue growth outlook. It extends to a step change higher in expected profitability as well. Okay. So let me bridge that for you. You've seen in our guidance, what we intend to grow for between fiscal year '26 and fiscal year '27. For fiscal year '28, we expect operating income of approximately $1.7 billion to $1.9 billion. And that is representing growth of approximately -- excuse me, 80% to 100% from fiscal year '27. Accelerating from the 50% growth year-on-year that we guided for from fiscal year '26 to fiscal year '27. And this equates to an operating margin range of 24% to 26% and as a result of further operating leverage in core and core AI, combining with new revenue beginning to scale.

As we've demonstrated historically, our model has the ability to convert incremental revenue into incremental operating profit and an attractive margin rate. Our overall financial framework seeks to optimize growth and profitability over the long term, compounding the overall value we're able to deliver to shareholders. I'll let you think through these numbers for a minute because even when I look at them, I cannot have a neutral reaction.

So let me put things in perspective for you because this is the same sort of mental model I had to go through myself. Fiscal year '27 is expected to grow from $3.7 billion in '26 to $5.05 billion in '27 at the midpoint of our guidance. So we are skipping the 4s, yes. We're skipping the 4s. Fiscal year '28 revenue outlook shows that we would be growing to $7.15 billion at the midpoint of the outlook from $5.05 billion. So we're skipping the 6s. Thank you, Steve. In terms of operating profit, we would be almost doubling operating profits in fiscal year '28, relative to fiscal year '27, right? That's where we stand today. And we will refine our outlook and turn it into proper guidance as we lap fiscal year '27.

All right. So -- let me summarize Chapter 2, the key takeaways of Chapter 2, that before we discuss the long-term financial profile. First, again, please understand. This is really important. Irrespective of the NAND cycle, we continue to gain share and deliver durable growth across all cycles despite component inflation or deflation. Second, our customers reward us with a value premium, our value premium, excuse me, that translates into high gross margins because of our differentiated software, which delivers them superior TCO. Third, our market opportunity is expanding significantly. AI is driving a TAM well beyond our historic $50 billion figure with a serviceable all-flash market expected to double from '26 to 2030 from $34 billion in '26 to $69 billion in 2030. Fourth, -- our preliminary -- excuse me, fiscal year '28 outlook reflects this momentum driven by accelerating revenue growth, operating leverage and scale benefits.

Ultimately, our confidence in the next phase of growth rests on a strong core, consistent share gains, cycle in, cycle out and expansion into much larger growth markets. This is why we believe that our profile is resetting durably for the long term. So let me bring that all together for you on what it means. The key takeaway is precisely that. Our financial profile is resetting to a higher and more durable level, characterized by sustained growth, increased revenue streams and expanding profitability.

So if you look at rule of 40 scores roughly for 26, we had a score of 33. And for those of you who follow our earnings transcripts, we did say that for the past 3 quarters, we hit rule of 40 already. And we also guided for the end of this year to a point where we're implying a score of 50 to 57, okay? That's where the 50 comes from. And for fiscal year '28, when you take that into account and everything I gave you on our preliminary outlook would be in the 60 to 70 range. okay?

For the long run, beyond fiscal year '28 I can go beyond 3 years. You can expect that we would be in the range between 50 and 70. The core assumptions in here is that our core and core AI continues to gain share. and that our new revenue from this new revenue category, I articulate for you is going to represent roughly 20% of total revenue by 2030.

In so doing, we want to make sure that we continue to grow the quality of our revenue and have more durable, more recurrent revenue with Evergreen//ONE. That's a very important part of our strategy. we expect our ARR CAGR for the next 3 years to be greater than 20%, and we expect our remaining performance obligation growth to be greater than 25%. As a result of our push in hyperscalers and Evergreen//One, one has to assume CapEx as a percentage of revenues to be in the mid-single digits. And in terms of tax rates, at this stage, and there's a lot of things we can discuss about tax, which is a topic that I absolutely love to discuss. You must expect an annual non-GAAP ETR effective tax rate in the vicinity of 20%. And do not model just a bit of advice, do not model tax on a quarterly basis. You get the equation wrong because you don't know the mix of our profits by geography. So what you have to assume for the tax rate, when you model it in your models, 20% more or less on an annual basis, okay?

And then we expect free cash flow as a percentage of revenue to track operating profit margin. No change. Now you're going to say, how do you intend to deploy the capital you're going to be generating. Our capital allocation framework focuses on 3 key priorities to drive long-term per share value creation for our shareholders. First, we got to prioritize investing in our core and strengthening the base. that is absolutely a must and will continue to do so. We will maintain disciplined forward-looking R&D-led investments to sustain our technology leadership and expand into high-growth markets.

Second, we approach strategic M&A with discipline. We are targeting tuck-in opportunities that offer differentiated technology, exceptional talent, or expand go-to-market capabilities. M&A opportunities will be evaluated strictly on strategic fit risk-adjusted returns, integration complexity and a fair price and a fair valuation. Third, after funding the business, strengthening the balance sheet and pursuing strategic M&A, we return excess capital to shareholders through share buybacks and withhold to cover purchases to offset stock-based compensation dilution. Also, over and above that, we may select to execute additional buybacks with excess capital. Hopefully, this makes sense. I'm sure you will have questions.

So this brings me to the end of my presentation. You've been very patient throughout the day. I want to thank you for listening and flying in from far someone, Andrew flew all the way from [ Montreal ], some of you from the East Coast, some of you on the webcast have been listening patiently. So many, many thanks to all of you.

I want to summarize this before we get into Q&A by saying Everpure stands at a pivotal moment, a pivotal inflection point. No doubt, Charlie will reinforce that. And we're backed by the strong core business, industry-leading R&D and premium margins. AI is significantly accelerating our addressable markets driving accelerated growth, increasing operating leverage and reinforcing a durable financial profile above Rule of 40 throughout fiscal year 2018 and beyond. As our cash generation increases, we are committed to be deploying that capital thoughtfully for the benefits of our shareholders. We believe we have the technology, the market opportunity, the financial business model and most importantly, the people and the culture to build a substantially larger and more valuable company over the long run.

Thank you very much for listening.

With that, it's now time before Q&A, I would like to invite Charlie, Rob, Bill, who is our General Manager of Hyperscale Solutions, Prakash, who you already met. And of course, we couldn't hold the Q&A without him, our Founder and Chief Visionary Officer, Coz, to join me on stage for the Q&A.

Wamsi Mohan, Bank of America. Thanks for doing all the presentation today, very helpful both from a technology standpoint and Tarek, yes, some really impressive numbers here.

My question is, firstly, you obviously articulated a very bullish sort of outlook over here, well about consensus, both for revenue and profitability. To execute this from a go-to-market standpoint, can you share what you might be doing differently, particularly on the hyperscale side? And within your guidance, if I could, are you anticipating incremental hyperscale wins? Or is this sort of like based on what you currently have announced? And maybe some color around that.

Sure. The go-to-market is quite different as one might imagine, between the -- everything outside of the hyperscale or hyperscaler is a very different go-to-market. It's really it's engineering lead, right? And Bill and his team really lead that engagement. It's engineering to engineering. Really, there is -- while we do have salespeople and they're very good. But honestly, when it comes to getting the win and closing the deal and all the complexity associated with the supply chain and contractual agreements it's really led by the hyperscale team overall. On the core side, the go-to-market is -- has substantially changed. We've made a huge investment in that area. It is far more sophisticated today than it was just a few years ago, everything from -- it's not just product and it's not just selling skills. It's really financial engineering that is working with the companies that we engage with, so it fits into their model.

Most importantly, it's -- and Rob sort of alluded to this. It's a difference between selling into a particular use case and now selling at the enterprise level as a partner where we are either 1 or 1 of 2 primary suppliers here. And we call that a franchise win. In other words, it's not a win against a specific workload or use case. They're choosing us to be their partner in data storage, right? And that is -- I would say that has matured over the last couple of years to a really remarkable level, right?

That gives us -- and as we gain experience in that as we gain more and more wins with franchise wins, we expect to see even more franchise wins. So -- as I said, while we really only get about 2 quarters of visibility in the core business from a pure pipeline level, on the other hand, we have years of experience in terms of gaining market share. And we see that gives us the confidence to say we're actually entering a new growth phase. Now you might ask why now? It's largely because we filled out the product line only about a couple of years ago, we couldn't aspire to really be a franchise winner inside a customer. We didn't supply enough. Now we do and with advanced value add. So I would -- yes, please.

I was just going to jump in and say once you do the second part of your question, I think, which is hey, does the FY '28 or beyond outlook contemplate or rely on needing a second or -- sorry, third or further hyperscaler win? I'd just go back to how we've looked at the multiple high-growth areas that we're now pursuing. Each of which alone has -- as Tarek has shown you, drives a tremendous TAM and SAM expansion. So in some ways, as we go and pursue and prosecute those markets, yes, hyperscale is one of them. We have multiple paths to drive that growth. We see multiple paths to get there. We have a diversified essentially approach to go do that without having to bifurcate whether it's our R&D or sales engine, it's good.

But I would say, look, '28 is based on today's business and the visibility we have into '28 doesn't require anything fundamentally new associated with '28. Of course, our expectations are that we will, in fact, at new hyperscalers. Our expectations are also that we will penetrate a greater proportion of their storage estate as well. Yes.

Jason Ader from William Blair. Just on the FY '28 guidance, Tarek, can you give us a sense of how much of that 39% to 45% growth might be coming from pricing? And then also, it was great to get that 20% FY '30 from new markets. Could you give us kind of a rough estimate of FY '28 from new markets. And then separate for Coz. Can you just talk about like what this journey has been like? I mean, you're a founder you're sitting here today, whatever it is, 15, 16 years later. Maybe just talk through how you're feeling

So actually, next week, we'll be our 17th birthday on Wednesday of next week. So I mean, it's been a great journey. Obviously, at this point, I'm having a lot of fun. I mean, S&P 500 is cool. What we're doing with the hyperscaler is really cool. The team, I didn't have to get up here and present. They got to do it for me. That's really good. I mean, our goal from day 1 has been to build revolutionary technology and really change this industry. And you've seen it from the start with all-flash and Evergreen. And you see what we're now looking to do in the future, which is go beyond just the storage and really revolutionize the way people can control their data and understand their data and everything we're doing around data privacy, data intelligence, helping people with the AI, we continue to want to do extremely revolutionary technology that will totally drive something different.

And it's a lot of fun to do. It's a lot more fun than trying to squeeze another dollar out of a boring old tech and things like that. And so we're going to keep doing that. And I think you're starting to see the results, and that's also very gratifying because all right, I feel we've got the best product. So I don't know why anybody buys anything else.

I would agree with that. That's for sure. So just to answer your question on the financials. The first thing to look at and to help you with in terms of your modeling. You saw about 6 points of operating margin expansion between the end of this fiscal year and the preliminary outlook, Jason, I want to use the word outlook, not guidance, right? So take this 6 points. If you assume about 70% of that -- 65% to 70% of that coming in from new revenue to 3 big categories that Rob has spoken about. Modern data software, scale AI and hyperscalers. Then you will -- if you assume a reasonable amount of OpEx going against these new revenue, you will see that our core is growing. And we are not assuming any particular pricing benefit associated with that growth. We continue to grow irrespective of where prices go. And I am never -- and no one in here, Charlie, no one in here will tell you that this pricing environment will sustain itself ad vitam aeternam. We don't believe it's going to be the case.

But our preliminary outlook factors in any scenario that you want on that front.

Hadi Orabi from TD Cowen amazing numbers. Tarek, if I look at the slide you showed FY '28, it shows like white bar and orange bars. Orange bars for new revenue and white bars from core. Just visually looking at it, it looks like the new revenue, which includes hyperscale and scale AI is approximately can be anywhere from 10% to 20%. And just like...

I tried to blur it. So you don't take a ruler and go and do the geometrical arithmetic to come up with that number. No, seriously. I'm not going to break that out, okay? One thing I want to leave you assured of and here's how new revenue is going to work. At the end of fiscal year '27 will give you a baseline of where we landed on new revenue. During the course of the year, we execute. Some of this revenue is lumpy in nature. So don't measure it on a quarterly basis. At the end of fiscal year '28, we'll tell you where we landed, okay? And that's how we intend to update you about penetrating these new markets. I think it's a simplest way, and you will also see through the expansion of total revenue and operating profit margin, where we get to.

Got it. And just a follow-up. You guys have 2 hyperscales already. Can you give us a sense of how much the FMs within the SSD capacity or non-HDD exabytes you guys have -- your share is by the end of next year, just to get a sense of how much the revenues from these customers can grow in the future.

It's -- we can't give you an exact percentage because I don't think we have full access to that type of information. But I can say it's still a very small number relative to their overall purchases. Yes.

Excluding HDDs.

Well, excluding HDDs. Although we think on a longer -- as prices revert to the mean, which we're not going to predict right now because we inevitably get that wrong, when -- as prices start to revert to the mean, then HDDs are open season for us as well.

Eric Woodring, Morgan Stanley. Thank you again for everything and congrats on the outlook you provided today. Just a quick question. As we think about going from fiscal '28 outlook to that kind of fiscal '29 fiscal '30 kind of rule of 40 framing, right? The low end of that 60 to 70 comes down to 50. I assume you're giving yourself some leeway there, but I'd just love to understand like when we think about that is giving yourselves some leeway on growth, is that margins just to maybe understand.

We're not predicting a slowdown.

Hey, listen, a rule of 40 at the 50% mark. So I said the way you want to, my friend. I'll be happy. Okay.

And maybe just one clarification, Tarek. Just on the 20% tax rate annually, when do we expect that to start? Is that fiscal '28 or...

So a great question. So great question. This is a very important point. For those of you who have read our 10-K, you will see that we have a pretty big valuation allowance. It was about $838 million from memory. As our profitability becomes more durable, more certain, we will have to release that deferred tax asset. And our current thinking is that we'll have to release it in the next few quarters. And so -- that's why I cannot be very precise today on the non-GAAP ETR. But we will be. And I think for the moment, the assumptions we gave you roughly 20% is what you can assume.

Howard Ma with Guggenheim Securities. Truly incredible numbers. I think we're all ready for celebratory trains at this point.

Wait a minute. Hang on. We got 6 months of results to deliver and another full year to deliver numbers in accordance with that outlook, Howard.

So Tarek, that's partly why the initial FY '20 guidance is so impressive just given how far out it is. And so my question is, -- what would you identify as the 2 main -- or the 2 biggest areas of conservatism in the long-term framework. And I'll share my thoughts, okay. Because I can't imagine you guys are assuming much recovery in volumes because those are limited by IT budgets. So if you look at top line pricing will still be a tailwind next year, but volume likely still a headwind. So the implied core -- the core revenue growth can't be too high. So you're kind of implying a lot of it is already coming from new product. And if you look at it from a margin perspective, too, you guys shared that COGS, that total COGS slide. And you're clearly not raising prices. And I still going to hurt your product gross margin, right? So you're not going to get that much margin expansion from the core.

So it kind of backs into likely a very strong -- a very robust new revenue contribution already and that -- but I can't reconcile that with the 20% number 3 years out. It almost would imply like that number doesn't grow much.

Well, yes, let me start with that. We are -- so elasticity, and we've learned this and Tarek showed the slide where we saw the volume and price elasticity in both positive inflationary markets and deflationary markets. And in each case, supply/demand, pricing economics work the way you would expect, which is when prices go up, -- on a relative basis, demand either doesn't go up as much or can go flat, but it's not -- it doesn't make up for the fact that overall, the dollars are higher. And on the reverse side, if prices come down, elasticity goes up. In other words, volumes go up, but it doesn't quite make up for the price decline from a total dollar standpoint. So they go in opposite directions. Price dominates versus volume typically. That's been our experience over 15 years.

So that's -- I think that's pretty well established. By the way, that's supply-demand economics. That's the economics 101, right? That being said, we expect to continue to pick up market share on a significant basis. if prices stay where they are now, we expect to pick up market share. That means both the volume and dollars. So we expect to continue to grow. If prices start to deflate, we expect to be able to manage that. Of course, it depends on whether they deflate or crash. But again, that's beyond our ability to be able to predict.

So current predictions are based on current pricing. We're not expecting any dramatic change one way or the other if that makes sense. But that being said, no, we are expecting the growth going forward to also come from core. But you'd be correct in saying that the majority of the growth is going to come from those new markets.

Asiya here. Just to follow up on the prior question. Just if you think about supply, I think that's been a topic. I think you addressed that in the last earnings call that you have good visibility, good supply, and therefore, you provided an upside to your fiscal '27 guide. Just help us understand like these outlook that you've provided? Like how much is supply concern through fiscal '30?

I would say that on the -- the new revenue side, which is primarily software. But to the extent that if a hyperscaler can't source or their suppliers, I should say, can't source, then of course, that would get us in trouble. We don't have concern in that spot in that area right now. On the core side, yes, the concerns around availability of supply have gone way down. Doesn't mean we don't scramble from time to deal with shortages here and there. But for the most part, that's not a, I would say, a major concern now. Now you talk to our head of supply chain, they might have -- they're being kept very busy.

But from an enterprise risk standpoint, I'd say it's far less than it was a few quarters ago.

[indiscernible]

I would say that it tends to be memory constrained right now. So there is -- but our volume relative to the overall memory constraints is relatively small. So I wouldn't say that's a big area of concern.

Yes, I would say -- I would say when we think about the neocloud solution with EXA, very similar to the hyperscale solution where the neoclouds themselves through their supply chains, procuring the hardware, we're providing mostly a software solution. And then, yes, relative to how we're thinking about the core business, no particular supply constraints contemplated in the numbers we put out.

Yes. It's Mehdi Hosseini here from Susquehanna . Two follow-ups for Tarek. First, you said we don't need to change the model and you provide an annualized number at the end of this fiscal year. Should we assume that the incremental or new revenues would be lumped into product revenue? Or to what extent -- if you could help us with the distribution of incremental revenue into buckets of revenue that you provided?

And then number two, you had a CapEx for -- to support the hyperscale. If you could just provide what that CapEx actually is and for what application.

Yes. I'll take the first part of the question and I'll ask Rob to point on the CapEx he will elaborate on more. So yes, the vast majority of the new revenue today is software term licenses effectively if you want to simplify it, no change to what we said around hyperscale revenue, it's license plus some non-NAND related componentry. It all goes into product revenue. And you may want to ask and I'll take the opportunity to answer the question. What would be the assumable gross margin for new revenue in total. I think what we gave you for hyperscale revenue, 75% to 85% is what you guys should assume.

And then on the CapEx side, I would say very similar to what we've done in the past in terms of developing our core technology around DirectFlash, but now think of it as qualifying multiple different vendors of flash multiple different sizes and chips. And so as we think about serving multiple customers and going to larger drives, you'd assume a slightly larger spend commensurate with what Tarek outlined.

[ Steve Cannon ], Morgan Stanley. I think I can ask a strategy question basically the same one I asked before, which is the storage is not the highest influence, highest power part of the enterprise data group, right? It's not best for us to writing. How do you get consideration. I think you answered the question very well technically. But strategically, how do you get in there and say, we should be part of your data stack. We're a software company. We're not a storage company. And how do you do that from that position of having your strongest relationships in storage. And so what's the go-to-market challenge there? What do you have to do in terms of hiring bodies? How do you change your go-to-market? How do you -- or is that dialogue happening at?

Well, it is happening, I would say, naturally, but it's also in transition. It's transitioning from -- and when I say it's transitioning, we have a set of value adds today that will continue. But now we're transitioning to a set of even higher value adds that appeal beyond the traditional storage purchaser. So today, it's -- it's an amazing amount of work and effort and labor and thoughtfulness has to go into managing a data storage estate today. One asked why. It's a pretty -- as you point out, it shouldn't be a bit more mundane than that, right?

And what we're doing with -- first of all, just making the product simpler to begin with, but then going into what we're calling Fusion and the enterprise data cloud simplifies their data storage estate and allows them to invest their time into higher level concerns. We truly want to make it automatic. We want to -- you don't go down into your basement and fuss with different plumbing in order to take a shower in the morning, you just turn an op that's what we want to get to with respect to data storage as well, right? They just turn it off, they get what they want.

But where it's going is your point, while the data storage itself doesn't have power, what we're really predicting is data is going to have the power going forward, even more than the application environment. Because without clean data, you're not going to get clean results, right? Applications are going to be -- are going to be much more focused on the workflow and not on controlling the data itself. And that's where the conversation is going. And that's where we're putting in a lot of our efforts around data intelligence around being able to better manage that -- once you have added context to the data, once you've made the data more valuable and more accurate being able to manage it according to the customers' policies rather than, again, fussing with the plumbing down in the basement. It's going to be -- we believe that, that gets into much -- well, we believe we are getting into much higher value conversations with higher levels in the IT world in this area.

And Steve, if I could add on to that. I think from my lens, there's no one single answer, right? I think there are a number of things that we have to activate across the entire go-to-market engine, well, frankly, the entire firm some of which are well underway, some of which were, I would say, in the early stages of. As Charlie mentioned, our historically typical buyer has been a storage administrator. That person, that individual may not -- may not resonate may not care too much about the value we're creating. And as I alluded to in my tone back when we were competing for workload by workload 500 KPO, $1 million PO, that individual could sign off on it. When we are now going and competing at a franchise level, putting 8-figure, 9-figure deals in front of customers, you can bet that's getting us conversations way higher up in the stack.

When you think about pursuing and articulating the technology differentiation strategy that Charlie just mentioned, that requires a new DNA in our sellers. That requires new DNA in ours SEs, our partners, the enablement, the training to go do that. So we're well underway with that. And I think just even most visibly, it starts with not getting box out of the front door, right? One of the things that really led to rebranding the company and changing the company name is making it very visible, very optically obvious that we are doing much more than just data storage.

So I think it's a number of these things that have to fall into place, some of which are well underway, many of -- all of which were put muscle into but it's all across the entire go-to-market engine as well as the product strategy and bringing that all together.

Yes, I'll comment. What largely gives us confidence at this point because, as Rob said, it is early innings. But when we looked at our customer installed base of One Touch versus per turn, we saw in the franchise wins, the large enterprise regulated environments, about an 80% logo like overlap initially. So we saw in the Fortune 500 strong synergies and we've gotten, at least in the early days, and I think Ashish covered this. The personas are like these franchise wins that are doing these big kind of fleet engagements with us are favorable and do like us. Our Net Promoter Score is high, and they are bringing the Chief Data Officer or CISO to the conversation.

So I think part of the leverage we have is just our customers like us. We've treated them well. Our Net Promoter Score is strong, and they want to do more business with us. Many years ago, we used to say that when we were a block only before we really got into too much file, the number of sales reps here that were just like end customers that pulled us. You only had file. We would -- we really want to work with you, right? And we're finding 5 to 7 years later now, that same reception in this within our installed base across the multiple personas. So the initial interest in demand is fairly strong across the access to the people we wanted to get to.

A couple of more questions here on the front, maybe MP, then Wamsi or -- sorry.

Matt Calitri at Needham. Thank you guys for having us and doing this. It's been a great day. Rob, I think it was in your section that you noted that hyperscaler interest is expanding beyond the top 5. What exactly does that mean? And how big is that overall pool of organizations before you start bumping up against like the neocloud opportunity?

Yes. Absolutely. I'll start and I'll tag Bill in as well keep them on his toes. So I think what I was hoping to articulate is the kind of core advantages we're delivering to the hyperscale -- the top hyperscalers with our solution set today, how we're packaging and delivering that, how that fits in the architecture. And yes, a lot of the dialogue we've had with the financial community over the last couple of years has been focused at the top hyperscalers. But it turns out, if you look at call it, clouds through 100, 500 to 1,000. Many of these firms, many of these environments design their environments, their architecture in substantially similar ways. So it would stand to reason that they would benefit from a lot of the same attributes. But Bill, I don't know if you want to add that.

Yes. So not only are they transforming their architecture to look more like a hyperscaler, they're also looking for supply chains that look like hyperscalers. They're buying at such high scale, they want to see this supply chain that they can rely on multiple vendors through 1 solution. So it's very attractive to them. And by the way, neoclouds are not distinctly different from that. They're acting that way, too. And the build-out there and with the frontier model companies is also they also want to look like a hyperscaler. So we're seeing that there's a lot of opportunity for the hyperscale solution outside of the top 5 hyperscalers.

Just to put it in perspective, though, I want to make it really clear. There is a very clear set of requirements on the whatever we call these companies, tech titans, hyperscalers, et cetera, for them to even qualify for our hyperscale solution. And that is they have to have their own storage software stacks. If they don't have their own storage software stacks, then we're very happy to sell them our standard product because that -- our standard product has those software stacks in it. It's only to the entities that up until relatively recently, we're only the hyperscalers that had their own storage software stacks, right? So there is a clear distinction.

[indiscernible] on behalf of Joe from J Morgan. Just wanted to ask, is there any meaningful difference in OpEx intensity for the core versus the new revenue opportunities?

What I would say is, you can expect the different OpEx lines, obviously, as a percentage of revenue, the percentage is to go down. But we will continue to invest in R&D. We have to invest in sales and marketing, less so in G&A, but we have to scale up G&A as well. But you can expect absolute dollars to go up for R&D and sales and marketing to get to deliver the kind of growth we articulated, we need to scale up across the organization. And there's no doubt some absolute dollar increases have to be factored into account into your models.

But overall, it's top line growth and significant OP margin expansion as durable resetting financial profile.

Right. If I might. On a blended basis, we showed you what it looks like. you have new product economics, whenever you introduce new products into new markets. And those new product economics are always generally lower gross and operating margins to begin with and with market share comes to margin at every level. So as we gain market share in those products, we should expect them to accrete to company average and then preferably you'll go beyond company average and start being accretive to the overall business.

Got it. And a quick follow-up. On the new revenue opportunities, I think in the chart, it looked like there was some contribution in fiscal '26 as well. Like was that meaningful enough? Or are you willing to quantify that?

It was. The answer is no, we're not willing to quantify that. It doesn't really matter. What we said, if you look back at our earnings transcripts is that we said we had tens of millions of dollars from 1 particular revenue stream back then it was hyperscale. But now you have to look at it as a combined category. The 3 are really where we're going to put our attention to. There's a tremendous opportunity in modern data software, a tremendous opportunity in scale, AI and also an equally tremendous opportunity in hyperscale solutions. So we'll give you a view, the '26 number is, to a large extent, not very important. What's more important for you is the fiscal year '27 baseline and how we're going to progress to the 20% by fiscal year 2030. Would you please pass the mic to Wamsi, who has been -- thank you.

Thanks for the follow-up. I guess just at a high level, would you say that adoption of AI is changing procurement cycles for you guys? And in terms of visibility, particularly on the hyperscale side, can you talk about if you have extended visibility and commitments in terms of exabytes or revenue? Or how are these commitments shaping now over a multiyear period?

I'll take the first one. I mean I think if we look at the core business, I wouldn't identify any specific change in procurement cycles due to deployment of AI, other than the industry CapEx deployment NCIs obviously created a supply chain pricing dynamic, which that has certainly -- and we've talked about it ad nauseam, certainly factored into enterprise procurement. But in terms of enterprise deployment haven't seen really any direct effects of that in the core business in terms of procurement cycles.

I will say in our discussions with hyperscalers, they and working with their procurement teams, they do share their long-term spend and long-term outlook versus AI. So that gives us good insight and ability to plan.

And '28 is -- we have firm commitments through '28. So on the hyperscale side and the other portion of 28 is based on continued market share gains and no fundamental -- as we talked about, pricing costs will change somewhat. We know this, but we're assuming pricing stays reasonably near where it is today.

I thank everybody is ready for a drink, I suppose.

Well, we can certainly break for a drink early.

It's 5:00 in New York, right?

Well, look, again, we really want to thank you all for traveling to see us today spending so much time with us being very patient as we got through the trading day to get to the meat but also really your interest and curiosity about how our business is built and what the fundamentals are really was quite -- we were very impressed with the Tarek and Rob and I were going back and forth with about how insightful many of the questions were and how knowledgeable about the market.

So I really want to thank you all for your time today and look forward to spending more time with you this evening. Thank you your questions.

Hello, everyone. Welcome to the Everpure Financial Analyst Meeting held for the very first time here at our headquarters in Santa Clara. We have a full agenda for you today. We will start off this morning with technology deep dives on data management and AI. This will be followed by a technology-focused Q&A session. We'll then break for lunch and resume again at 12 noon with the main session featuring Charlie, Tarek and Rob. For anyone needing information on things like restrooms or WiFi, we do have people in the hallway who can help you with any questions.

When we return from lunch, Charlie will lay the groundwork with an overview of our strategy and our expanding opportunity, actually our significantly expanding opportunity. Rob will cover our fundamental advantages and our right to win in 4 high-growth market areas. These include our core plus 3 additional high-growth market areas. Then Tarek will conclude with our foundation for future growth and our long-term financial framework. Following that, we'll conduct a final Q&A session with Charlie, Rob Coz, Tarek, Bill and Prakash.

And then after the final Q&A session, for those of you here in person, we'll have an informal cocktail hour. Before we get started, I'll remind you, as I do every year at this event that the value of everything that we do at Everpure is driven by our software. This includes our core business as well as our 3 additional high-growth market areas. And for all of our high-growth market areas, once again, it's always about -- thank you. the software. Okay. So now I'll turn it over to Charlie for his opening welcome.

Good morning, everyone, and welcome. And for those of you who are here in the immediate audience, thank you so much for coming out. We know that it's a long trip for most of you and many of you, and we really do appreciate your physical presence here. And I also want to thank everyone who's online who will be going through our discussion today.

So we're very excited. The team here is quite excited to speak to you about everything that we're doing. A lot is changing. Obviously, a lot is changing in the outside world. A lot is changing in our industry, but a lot is changing at Everpure as well. And it's a great time for us to be able to express this to you in many ways.

One is we believe that our business is accelerating. We hope to be able to show that to you today in numbers, but it's also expanding into a number of areas, and that's obviously what Paul alluded to and as you can see from the agenda -- different areas that we'll be providing you a lot of background in. One additional thing for those of you in the audience that is here physically, we left you with a paper on the table, it is a bit of a long read. It's about a dozen pages, but we think it's a seminal view of how IT architectures are going to fundamentally change over the next decade.

We believe that this brought on by AI it is driving the need for organizations to start to organize around data rather than around applications. And that's a very fundamental change in how IT architectures are built. And that fundamental change really is going to open up very new opportunities that we as Everpure are pursuing and other vendors are going in that direction as well.

But unless you understand how the -- how -- where we are today and where we're going as an industry, how that's changing and moving towards a different architecture and a different relationship between applications and data you won't fully be able to predict all the changes that take place. So I highly recommend that in your leisure time that you scan that scan that document and get a good understanding of it.

With that, I'm going to invite our first speaker on the stage, Prakash Darji.

All right. So I'm going to double-click on the first area Charlie mentioned, which is this idea of Davis centricity. We call this data primacy. -- it's an inversion of the application architecture. So as we get into our business, we're really talking about how we can help customers maxims the value of data, wherever it sits, pure non-Pure. We're getting into the focus of managing the value of the data and getting the most leverage out of it.

Two, why we believe we can do so is we already have the trust of lots of mission-critical data across a customer's landscape, both structured and augmented with unstructured as well. And the third thesis we're going to get into here is why a storage company has to do that. Because as you think about data, why this context need to be both traveling with the data and be persistent in storage. So if you think about the biggest data management challenge today, and I bring this back to prior to here, I worked at SAP, 13 years building applications. both financial and database.

But a large portion of the world today is customers buy applications and IT departments integrate the data between those applications. And if you think about the total spend, actually, 80% to 90% of all of the spend that goes into the integration work between all the assets you have. This compounds the bigger you get. And largely, the data inbration costs are largely the biggest costs associated with running an enterprise today. And if you think about agents, and agents access to data where it sits, where context is trapped in applications or in Lake houses, et cetera, that actually is going to blow up and be more exponential in terms of integration cost.

So these silos of information that sit in applications, warehouses, et cetera, need shared content to do anything with AI and agents work better directly on data. So that is a challenge we're seeing in the industry today. Now if people are running applications, you'll see SAP strategy saying, "Hey, you want to approve sales orders, so bring all your sales force data and SAP and Salesforce is saying, "Hey, you should build use agent force and bring all your SAP data in the sales force and data Brooks is saying, okay, bring all your data into data books.

Everyone right now has a part of the context picture. And this is what led to the fragmented context across the application. So our first approach is how do you create shared context with data wherever it sits. So leverage the data in place and understand how to build shared context. Two, instead of moving the data between systems, how do you allow reusing the data where it sits? And what that does with the integration complexity is it inverts the way you develop apps.

Historically, your data has been owned and encapsulated by the application, right? This is my finance data sitting from my finance application. This is my sales data sitting in my sales application. We do see a world in data primacy where application workflows plug and play and we use data across multiple apps. And it is a different way of building an architecting.

In the early days of SAP, we -- like when I was there, we were trying to build a big schema to handle what we would consider an tender. And largely, that became siloed and fragmented as people started buying different applications and not consolidating to a single referential data model. In the world of AI, we do believe that this idea of plug-and-play workflows will allow you to reuse data where it sits.

So when you hire a new employee, you don't need to copy all your financial cost center data, I can just update your financial costs. So the separation between HR and finance should not be as wide as it is today. So let's get into it. Data primacy. If you take this approach, it reduces data integration, reduces your data copy, increase of the data coherence, meaning the meaning of your data across your landscape, obviously, less is always better in a cyber attack surface, right? And doing so, you can also optimize your token spend, meaning that inputs you have that are relevant to models instead of building a large language model factory, an SLM with the right data in the right context, will get you better results at lower cost.

And we've seen that over and over again, and that's how agents are being built. They are being purpose-built with targeted data. So our beliefs as a vendor. First, we believe data will be distributed. There won't be one thing creating data on the planet. You will have sensors and you will have edge devices and you'll have mission-critical applications and innovation coming from everywhere. So we do believe that's to continue to be true.

Two, we do expect over time application development to change, a lot of application development was typically written in source code. I would write my pricing in source code. That was part of how you would do intercompany consolidations. We believe that will be moved to more configuration based as data. It will be persisted at data. You're already seeing people do this as markdown files and Claude. Kubernetes actually started this -- and Doran Kubernetes development, people actually started moving configuration into configuration files and compiling those configuration files. It was kind of the first view of what I would consider data primacy for software engineering. It moved to that approach where you had configuration that you could update and then compile.

Now context will be stored with, this is interesting because if you think about data and storage specifically, has limits. It's like what physically fits in this data center, what physically fits in this location? What is the performance of this thing? So over time, as data expands, you typically have to say, okay, I'm going to expand it or I'm going to place it over there or move it over here, but every time you do so, you're , I would have to retrain.

So if you really want to preserve what you've learned about that data, you actually have to persist that context with that data. And largely, when you have that context, it's not about the context of what's in that file. It's not -- this is a PDF with an invoice number. But I need to know where that invoice is being used across other datasets if I'm loading data into agents to do invoice approval. So shared context is needed.

And what this means is a storage operating system then becomes a place that need to store the data context and provide shared context. It's a persistent memory problem, right? Over time, a storage operating system is the only way you can actually persist memory context for data. And that's why we've decided to enter the space and we look at why storage and data need to come closer together.

And if you think about the problems, there are different problems we've historically dealt with, a storage problem is typically oriented with data residency. And when we talk about residency, it's not just a physical location. So where does it sit? What performance tiered or performance class do I put it in? What protection group do I put it in? So when I was an application developer, I always thought, okay, where am I going to place this data? So I just thought about placing it in that thing.

And I would inherit the policies of that thing Chad is going to get more into that later, right? And what cost tier is it placed on? You've heard us talk about tiers of cost. A data problem is different. And in the world of AI, data sovereignty is who owns your data, who's allowed to train on it, who can use it and does it still differentiate you? So if you really want to understand you need to go look at the data and have policies around managing the data and data access patterns as well. Because the more valuable AI becomes the more control over the underlying data you need. Otherwise and I bring this up many years ago, Walmart had a policy saying, "You know what, we're not going to use AWS as a cloud infrastructure provider. Why? Because they're a competitor in retail way back when.

And largely, there was a fear that they would train on their data. And if you see what's going on right now, if you don't control your own data, and maintain the ownership of how it can be leveraged, all of it is being used to train other people's models and IP. And that's what Data sovereignty is about. So what's fascinating is the market, the traditional competitors are actually working against data sovereignty.

You see a lot of vendors, hyperscalers, app vendors like SAP sales, Oracle, Databricks, different vendors all taking a walled garden strategy, bring all your data to me, and I will provide value to you in my platform. It's about data collection. Our view is you need to use the data where it sits and you need to allow the customers to own their own data. Now largely, we've entered and we -- this segment with a set of capabilities for data management and there are other vendors that could take that strategy, integration providers and iPaaS vendors actually work with the data wherever it sits.

But they're not focused on this data management problem, they're actually chasing agent studios, agent orchestration, data contextual, data versus the data contextualization we're talking about. So it creates this opportunity right now where we see this inflection point of a problem that needs to be solved and not a lot of people solving it. So I'm going to bring up Ashish and he's going to talk about how we're solving this problem.

Good morning, everyone. Everpure Data Intelligence is an important expansion to our portfolio and our offering. Historically, we govern the data and store the data for our customers using our unified data ever pure data intelligence adds universal data intelligence across any kind of data. We discover data regardless of where it sits, whether it's in the cloud, on-premise, even on mainframes. We then also think about it from a structured and unstructured perspective because data is data, and it needs to be discovered.

And we look at it whether it sits on Everpure data storage or on any other storage as well. This allows us to then classify this universally accessible data in such ways that we can determine what's the sensitivity behind that data and how you should manage it. And that's when the magic starts. We start to contextualize all this data based on business processes and the relationship between data, whether this is an employee, whether it's a customer, whether they're both of them.

And this allows us to provide this contextual intelligence that's required for AI tools to have better inferences and much more accurate decisions that they can provide to their various customer business use cases. All of this will be available at full scale on the Everpure data infrastructure and is already being used across multiple different organizations today. Now I've spoken to about 100-plus customers in the last 3 years. And there are 3 very important themes that are coming up.

First, customers want to have context-rich data and intelligence about their data right at the source so they can make good decisions based on that data. Second, we are seeing that large enterprises are bringing a consortium of decision-makers when it comes to purchasing data management tools. This includes the CISO and the Chief Data Officer as well. And as a result of this, the budget is tripling in terms of their spend on data intelligence tools because the CIO is using the information, specifically to understand how should the data be managed in the infrastructure.

The CISO is taking and prioritizing that intelligence-based information to see what's the posture of the data, what is the sensitivity of the data to take action on the data and the Chief Data Officer is being added to the mix because they can now provide actionable intelligence based on AI-ready data, which has context, which has relevancy and which allows those cost efficiencies to come together on AI tools because it's bringing the most context-rich relevant data to pay.

So this allows us to have not only one scan to enable multiple different use cases, but 3 different personas also get the benefit of it. This is also the reason that this market for data intelligence is going from $6 billion today to $21 billion by 2030. As I was saying, this is used by Fortune 100 companies today by large enterprises across the world, everpure data intelligence is delivering great results for our customers at a major credit card company they scan 14,000 databases in 2 weeks to really understand what data landscape looked like because they were having data subject access requests coming to them. It used to take them 21-person weeks.

Again, I'll repeat that, 21-person weeks to respond to each data subject access request, which is a GDPR requirement. With Everpure data intelligence, they can do it in less than 3 minutes. Massive cost savings for this credit card company. Similarly, another payment processor found 91 million records of credit card information stored in a completely unprotected database. That could be a huge security risk. We were able to identify that.

And to add to it, we were able to tell them that instead of what they thought that they had $189 million personal identifiable information. We could reduce that down to $39 million, largely because they had a lot of duplication of the data that Prakash was talking about when you do integration, you start to copy data and make it available outside of those applications, which creates a duplication.

This reduced their cybersecurity costs tremendously and was allowing them to also be more efficient as a business. And lastly, a global large bank was asked to comply to GDPR restrictions and policies and in 2 months, they were able to scan all their data and get ahead of the auditor request by providing full GDPR capabilities because we were able to provide them very specific visibility about their data, not only visibility but also the context that's required to make better business decisions.

With that, I'd like to invite Prakash back up.

Thank you. So for those who don't have the history, this came from our acquisition of One Touch. Ashish was the CEO of the company. He's now our General Manager in running this business for Everpure. And largely, the history started in more of a CISO security posture area. But that leverage he mentioned around users expands because now in the world of AI, if you add a catalog of where your information is and you wanted to say, "Hey, I want to build that I think I used that example of invoice approval.

What if you could just query with MCP, where's my invoice data and what's primary source? because otherwise, a, I might think the spreadsheet on someone's desktop is the same thing as your back-office NetSuite system. It doesn't know. So you absolutely need to catalog discover and not, and this relevance now is more applicable in the world of AI, so what's interesting is we have a lot of leverage that allows us to work on the data that matters most.

For -- ever since I've been here, we've had a strong Net Promoter Score keeps building every year. 64% of the Fortune 500, 84 Net Promoter Square, our customers trust us. We have strong trust in serving mission-critical data. Mission-critical like your most mission-critical financial records, transaction, critical databases, SAP, Oracle, et cetera, run on our FlashArray today. .

And with our data plan, we've extended to support block file objects structured and unstructured. So we can bring together serving this data to our installed base in a turnkey way. So we have a massive leverage point on how we provide context directly to customers today. Two, that leverage compounds because as people look at data intelligence and run these scans of their landscape, we can now see our share of wallet into accounts and show customers the value of moving data to Pure. That provides a product-led growth opportunity and ability for us to expand our footprint into accounts.

Two, as we take a look at our storage system, by having the data intelligence embedded and having this context and Chad is going to get into this in a little bit more detail, we have the ability to go ahead and say, you know what, I could go ahead and provide database management for the storage system. Because today, in storage, if you wanted to say I've got this invoice approval agent, how do you offer high availability? Has anyone even talked about that? It's mission-critical. It's improving revenue coming into the company.

But a storage system is where you set policies for RT-00,-RP00 recovery points, et cetera. So if you wanted a high availability, you would need to provide it using the storage system, which means you need to manage your policies based on the data itself, not the storage. And we're unique in the approach we're taking care. So whether you're a CIO and you start with the journey of storage where we talk about running workloads, automating configuration delivering performance and capacity SLAs, you end up at a place where you're optimizing for data residency.

If you start as a Chief Data Officer, we're talking to people about, one, the first step is total visibility discovery or data, understand it and reuse it. And for CISOs, this gives you a unique view for cyber resiliency as well. So you have the ability to not just look at the data, but to verify it and recover clean copies. So use cases, we see customers wanting to build secure data on claves. How do I ensure all private information or PII resides in the zone and doesn't leave this soon. It's a capability where storage and data management need to come together, ransomware use cases where not only do you detect, but how do you verify data integrity.

You need to look into the formats of the data, which data intelligence provides. We talked about the agent policy use cases. And in AI, how do you get better quality of outcome at minimal token cost? Well, one, only bring the relevant inflation into our model because your input to output token expansion comes down drastically and your results go up overall.

We've seen that proven over and over the early problems you saw in AI where people actually just you hear the word AI work slop. There was a lot of slop going into AI models, saying, just give it everything and ask questions, and you'd see a lot of things in questionable quality. So focusing on the data in is important. So with that, I'm going to bring Chad over, and he's going to double-click into some of the core platform. So thank you.

Thank you,. All right. Good morning. Excited to showcase some of the innovation that we're building into the core platform today. Prakash and showcased a lot around data intelligence. It gave the benefits of being able to discover, classify and contextualize data and provide that context that's needed down into the infrastructure stack that we're building around this core platform to allow customers to take the innovation that we built in and just continue to compound that value that customers have.

You'll see in this presentation, I'm going to go through 3 main areas: These areas are the areas that customers are most challenged with today, and we spend most of our time digging in with them to help them along this journey. The first step here is helping them being able to unify on 1 core platform. The fragmentation of data has become a fundamental problem and shown its cracks with AI.

The second big area, as Prakash talked about, is the ability to be able to control data, apply policy directly to the data and allow it to follow along as it moves within the environment. And then lastly, we're going to talk about the ability to operationalize AI, optimize it, activate the data and make it so it reduces the overall token cost. So we're going to dig into all of these different areas today.

I'm excited to spend a lot of time with you on just kind of a lingering and the differentiation in comparison to what we're seeing in the market today. So when we look at 1 core platform, one of the biggest things we hear from customers is, I've got way too much stuff to manage. Silos exist all over the place. Each one of those silos have their own UI, their own console, their own ad hoc configuration. It's a mess.

Every single application gets deployed storage requirement creates a new one. In the world of AI, you actually need to be able to access all data holistically, not just be able to look at 1 special system. Our competitors, comically enough will tell you, let's go build another silo for AI add more complexity, create new copies of your data and have you manage even more challenges that are out there. Each of these separate systems are their own snowflake, it's very hard for people to be able to manage their data estate without them being able to actually unify this together and apply governance directly to the data and take the context that we're taking from a data intelligence and apply it down to the infrastructure stack.

So we decided the model needed to complete rethink. The rest of the market was thinking down here. This is storage management. Now we made the best boxes on the planet that produce storage solutions a lot of great software to make them simple, efficient, reliable, but most people are continuing to stay down here. And our goal here is to try to move people up the stack into building better efficiencies and workflows that allow them to be able to really get value from the system.

This first stage is about providing autonomous operations. It takes intent from an end user. What do they want from that data and allows the infrastructure to actually manage it. It's great. Think about this model for a quick second where you're able to just define what do you want? I only want models to access this data. I want to be highly resilient and not have to manage any of the configurations to do so. On top of this, we're building in data management componentry, which we talked about earlier in data intelligence.

This informs our intelligent control plane and unified data plan, which I'll go through a little bit later and what it needs to actually do based upon the context of data because just executing intent that's going to always allow you to understand exactly what data and what kind of context is associated to it. So to fix this problem, we decided to build what we call an enterprise data cloud. What an enterprise data cloud is as it starts with an evergreen architecture. It's one of the things that customers love the most about Viper is the fact that they can get new generations of hardware with no disruption -- it's a subscription to innovation where they consistently get more and more value. It compounds every single year, and they never move off the systems whatsoever.

We then built a unified data plane. It allowed them to break through all of their silos and provide one common user experience throughout but do so with multidimensional performance so that they could handle every possible workload from all the way on to AI. On top of this, in order to create this unified data plan, we created an intelligent control plane. This is where the brains of the operation exists from an infrastructure stack. It takes the context from data intelligence and now applies it down through the infrastructure core, taking the intent of users and applying it directly to the infrastructure stack.

And of course, as you saw with Prakash she's talking about data intelligence. This just becomes the context and meaning. All of this put together is our core platform we deliver to customers -- and the value that they get is substantial. We've put so much R&D development efforts into this that they have changed the way that they think about their model, and now they're moving from storage management into actually managing their data directly.

Let's dig in a little bit on what a unified data plane is. I get a lot of questions on what exactly is this? So the first is that we delivered 1 common operating model. There's no different consoles or UI. It's 1 single experience that users have across their entire data estate. Second is we've built this on a core fundamental technology, our key value for that provides metadata optimize access to data so they get a very common experience across each 1 of their use cases. On top of this, we built this on an evergreen architecture.

So unified data plan can't take downtime. It can't take disruptions. You can't continue to be moving it. It's got to be solid as the foundation that this is built on. Inside the data plan itself, we've got a whole bunch of cool technologies. FlashArray does really well for transactional latency environments such as databases and virtualization. FlashBlade is great for scale-out, analytics and AI, where high concurrency and high throughputs needed. For those that want to scale well beyond what an enterprise does into the Nel cloud scale, we have FlashBlade Exa that we've built for tens of terabytes of performance.

And then for our hybrid customers, we can go across cloud and the enterprise to build a solution where that data plane connects all of them. The benefits here is that we can apply block file as well as object capabilities across enterprise applications and also AI in the exact same platform, removing the need to create new silos and new complexities for customers. After Green architectures, I think that everyone has loved the most of help here from the very beginning. You never replace the gear. It swaps out modular in nature, stateless, so it never takes a downtime.

And over the last 15 years, we've delivered 9 different generations of hardware for customers. And every time they get performance gains capacity density gains, all without taking disruption. I use the analogy of imagine if you wanted to remodel your house or even move to a new house. Imagine while you're out to dinner with your wife or significant other, that your house when you came back was brand new. Didn't move anything moving sucks, right? You've got to break some stuff, you lose some stuff.

But in a world where the house just gets better instantaneously is the experience we deliver to customers. And on a unified data plan, this becomes incredibly important. Our software innovation continues to compound value for customers because of the fact that we deliver about 200 net new features to customers on an annual basis. So they're getting the subscription to innovation consistently, which adds massive value.

Our direct flash technology was built originally in the enterprise, but now we're delivering this in hyperscale environments, and we're getting testing now at scale that no other vendor has ever seen before. The value of this is resiliency. We get 10x better resiliency than what you'll see in off-the-shelf SSDs overall densities at 300 terabyte drives will be 10x more than what our friends in the industry around this for RAC and then also reliability and operating costs become key capabilities of continuing to scale this.

And the reason why is because we built something so different that the rest of the market continue to just use the off-the-shelf devices. It was software-driven, globally managed, and we made it so it was highly efficient and resilient from the get-go. All right. So now that we've unified the data plane. Let's talk a little bit about what we're doing in the intelligent control plane to actually apply this control mechanism around data.

So the first thing is, is Fusion actually fuses these systems together and makes this unified data plane, but it also becomes the control mechanism for managing data. What we do for customers and what's great is we have telemetry data that comes to us every 30 seconds from all of our systems to our own cloud, and we can help them be able to define what intent should be based upon past experience. What's intense understood and even better when you actually plumb in some of the data intelligence capabilities, you get the ability for the platform to actually be able to execute autonomously and the way this works, as Prakash talked about, is it is applied directly to the data so that the system can understand the intent that the end user has based upon our contextual knowledge and it can monitor the system to ensure it's delivering on the commitment they gave back to the business.

This ability here not only increases value to customers, but just think about this for a second. You don't have to control anything and manually manage it. You apply intent that goes across every single one of your applications, and it autonomously runs without anyone managing it. The benefit that are huge in the enterprise who is dealing with, in many cases, hundreds if not thousands of storage systems at once.

We take this context and in many cases, it may be storage context, such as what Prakash talked about, data protection or resiliency or residency, but also you'll see here shortly, data intelligence context and we apply this to a grouping of data, a semantic data group, we call it. This allows you to have datasets that are organized now by meaning not just because it happens to be sitting on a storage system -- this meaning thing allows you to be able to define policies such as this dataset must remain in Germany or redo models that have defined can actually use this data set.

This intent-based environment is so ridiculously powerful to customers. But what's more important is the ability to autonomously execute the capabilities within our intelligent control plane. We use this concept called Super context. It's a mix between storage context, which we know from the storage infrastructure space, paired with data context that comes with data's meanings and relationships and the like, the power that these 2 bring together allow people to be able to have better governance more trusted AI and more accurate autonomous decisions.

It allows you to control the data at a data set level but now actually understand what it means to the business so that when you apply intent, it's not just make it more resilient. It has an awareness to the business intentions associated to that data as it moves within the environment. What's been great is Fusion, which is the capabilities that underpin the unified data plan applies much of the intelligence that we've built into the control plane, so that users can take intent and apply it autonomously has been well received by our customers.

We already had 2,300 customers that have adopted the technology, and that has increased almost 2x just within the last 6 months. Customers are loving the fact that they can take this technology and remove storage management altogether and autonomously have the infrastructure manager itself while giving them the opportunity to truly understand their data above within data intelligence.

Options IT, as an example, went and saw massive savings, hundreds of hours saved, in fact, per month are users because of the fact that they weren't actually managing storage anymore. They freed those engineers to do more strategic aspects in the business, things like looking out what data they have out there, how they start making it more fish and getting it available to get more and more value from the data.

Evergreen One takes a very similar concept. It's our storage as a service offering that takes true SLAs. It's the most differentiated offering in the market today because of the fact that we commit to these SLAs to end users, this intent follows all the way through to consumption. Consumers can say, I declare my outcome. I want a certain performance range. I want a certain resiliency. I want a certain efficiency factor and it will deliver that to end users as an SLA that the system is to holistically. That then as the user scales, it allows them to be able to pay for what they consume as they continue to grow.

And if you pair this together with Evergreen as an architectural approach, as we talked about earlier, Evergreen allows you to remove the need to worry about the technology. Think about this for a second. If you get upgrades to the newest generation hardware performance gains, density gains every 3 years as part of this subscription, you don't have to worry about what technology is going to evolve to over time.

You can take the guesswork completely out of it. Evergreen One, though then takes the guesswork out of the actual consumption itself. And so you don't have to predict the demand anymore. You can just have it actually scale alongside that environment, reducing the complexity and, of course, risk that comes with procuring systems as a whole.

All right. So now that we've gotten the unified data the benefit to how we're applying intent and economy.

Let's talk about what we're doing to optimize AI workflows to build in better efficiency and truly activate the data for our customers. So the first thing when you think about it, the nice part about a unified data plan is you're running enterprise applications on that exact same infrastructure. We talked about this earlier. We don't think customers should create another silo, this should run on the exact same thing because likely will run in the same data that is already running the enterprise itself.

But what's even more key here is that multidimensional performance within the unified data plane allows them to be able to run AI workloads even more efficiently because it's both on the SAM infrastructure stack. It has the same policies that can be applied -- and it doesn't require additional copies or silos to be built. Second is we started to figure out ways for us to activate data. The first thing that most enterprises look at is how do we take the data that I currently have today and make it so that it can be used for retrieve augmented generation.

So that when I create prompts, it can actually give me responses based upon the data that is true to my enterprise today. What data stream does is it allows us to activate the data. It vectorizes the data, discovers where those data sources are out there, provides the vectors as part of the retrieval process, so that it can actually access direct enterprise information and be able to apply governance directly to those vectors as well.

When it comes to efficiency, we realize that the tokonomics and optimizing for tokens is going to be a big focus so that people don't continue to generate off the same exact series of pros. And so what we did is we created what's called key value acceleration inside the product, which goes and keeps a history of prior generations. -- which allows for faster inference, 20x faster inference and even more so in some testing that we've done. It allows you to reduce the amount of GPU utilization per run and inference, which allows you to utilize the GPUs more often, and it optimized for overall token cost, which becomes a huge benefit we're actually leveraging much of this internally as we're building our own AI stack to allow us to be able to increase efficiencies in the way that problems come in.

Certain prompts don't need frontier models that are sitting out there. some can land directly on open weight model sitting locally on our own systems, our own data. All right. So we talked about quite a bit here. We talked about 3 main areas of focus for us that are helping to drive significant value for our customers. The first is around unifying the data plane, which gives us the ability to continue to grow the platform, continue to add new workloads and be able to take AI and run it directly on the same datasets, which just drive more and more compounding value to our customers.

The second big area is talking about scaling through software. We're taking our software capabilities with infusion and our intelligent control plane and allowing us to actually help govern data and control it. And this new world of AI this becomes paramount because we want to make sure that policy is right next to the data so that agents are aware of the intent and understand whether they should or should not access it.

The last part is around optimizing the data itself where we try to improve AI economics. We enable users to get faster inference, activate the data that they have today and reduce the amount of duplication that's sitting out there. All right. That's a quick update on just kind of the core platform.

I'd love to invite Rob Lee up next to go through SCALA. Thank you. Awesome.

Well, good morning, everyone. Great to see a lot of familiar faces. I'm going to pick up where Chad left off and bring a wrap to the morning's technology sessions by diving a bit further into what we see happening in the AI space. Chad talked a lot about where we see AI being deployed in the enterprise. -- how we're meeting that need. We'll talk more about that in the afternoon. I want to switch gears here a little bit and take this up in scale by 10 by 100x.

Talk about how we're serving the needs of the neo clouds and how we're doing this with FlashBlade, which is our flagship product in the space. So if we step back from this and we take a look at what we see happening in the AI space, a couple of things, right? Certainly, it's growing. We're seeing also now a mix of new workload types coming to bear. And this makes sense. Years ago, it was all about training, about building frontier models, building who can build a better model with more parameters as those models develop as the harnesses develop, we're now seeing a much broader mix of use cases being deployed.

Yes, the training, but also then putting those models into production in inference, in agentic deployments, so on and so forth. Also, what does this mean? This means now we have continued explosion of AI deployments. You see this in terms of GPU deployments. You see this in terms of the high-performance storage attach that goes with that. But you're also seeing a much broader spectrum, if you will, a much broader dynamic range of different workload types coming to bear. It's not just training anymore. It's not just 1 workload type all of these things coming together and there's so much of it that the enterprises can't house this internally themselves, right?

They can't deploy the GPUs themselves. They're not going to. It's more than the hyperscalers can deploy. And so what you see is you see the neo clouds, right, filling this gap, right? And so what you see then is the Neo cloud serving as a point of concentration point of consolidation for all of these workloads all coming together into one. A lot of coverage around the neo cloud space in the market. A lot of the focus in this coverage has been around GPU deployment, who can get the GPUs, where is the supply? How much is it going to cost? Can I get the power? Can I build the power plants in the buildings and data centers. That's super important.

But actually, for the neo clouds, what's more important is driving GPU utilization, right? GPU deployment is about spending money. GPU utilization is about making money. This is about how they charge for GPU hours, tokens per -- whatever the metric is. And -- here is where storage has become -- has emerged as a significant bottleneck. The traditional purchase of storage have emerged as a bottleneck to driving GPU utilization.

And that could be because of performance limitations. It could be because of the ability for traditional storage approaches to be able to adapt to the mix of workloads now between training and inference, and agentic deployment could because of the lack of ability to adapt to the multiple different types of data that's being worked with. It's not just LLM and tech anymore, you're mixing this with images and video, stability and reliability, right? That's the easiest way to drop your utilization is if your infrastructure isn't available, isn't reliable, then the whole thing stops working.

And so as we've engaged in this space, as we've worked with Neo clouds, this has emerged as one of the biggest challenges that they have to face once they deploy the GPUs is, hey, how do I make these things actually utilized to where I can go and build for them. Interesting data point, there was a third-party study done that showed that across the board, about 85% of GPU -- sorry, AI clusters across the board run generally below about 85% GPU utilization because of storage bottlenecks or reliability or availability issues.

So -- why is storage so hard here, right? It comes down to a couple of things. Number one, there's a lot of data, right? Whether you're talking about training and massive datasets, whether you're talking about inference, which produces and consumes a lot of real-time data whether you're talking about agentic deployment which certainly produce and consume lots of data but then also need to be audited and governed, -- there's a lot of data moving very, very quickly, and it all has to be located close to the Chad just talked about how the last thing you want to do in the enterprise, much less if you're a cloud is build another silo just to serve this data need.

And so that's a big challenge. When we look at the type of data, I talked about the mix of workloads that are now coming to the forefront. When we look at the type of data that's being worked with, it's also changing. It's no longer just text prompts, hey, help me play my vacation. It's correlating that with audio, with video, with images, multiple modes of data all being worked with at once. And this is where a lot of the traditional approaches that have grown up in the traditional HPC or high-performance computing era really fall down, right? If you think about traditional HPC, this was a set of technologies developed for particle accelerated labs and nuclear simulations, right? Things that you plan for 2 years to do an experiment. Experiment runs for a day and then you're done. -- simulations.

They happen in human time. When you think about how a Neo cloud runs, you've got training. You've got inference. You've got agentic workloads all running for multiple users all simultaneously. This stretches traditional high-performance computing approaches and products well beyond the it. Reliability, right? Sometimes people forget that Neo clouds are clouds, clouds have SLAs, they have obligations to their customers. When the infrastructure doesn't work, those SLAs have penalties, much less not driving utilization, there is significant financial penalties.

And so these are operations that run extremely lean. They don't have large storage teams. They need stores that's just going to work even in the face of incredible, incredible demands. I talked a lot about how the variety of different workloads is driving different types of data access demands, but there are some workloads that just need absolutely mind-blowing performance. We see this in training, large-scale training environments with checkpoints.

As you're training these large, large models, this could take days and weeks you need to take checkpoints along the way to make sure that, hey, if there's a blip, you don't have to repeat that whole process that ends up driving a ton of bandwidth requirements. So you've got a wide variety of performance needs all colliding into one. It's all happening at the same time. It all has to be super reliable easy to operate. And oh, by the way, it needs to be secure and governed and compliance.

So there's a lot going on here. So what do we do? Well, let me start with what we're not going to do, which is build a brand-new product from a clean sheet of paper to go solve this because as it turns out, we're working from a solid, solid product and technology base that allows us to go meet these needs by repurposing and repackaging a lot of the core technology we've developed over the last 15 years. We start with FlashBlade, right? And we take a look and we say, how do we scale that up to meet these needs 10x, 100x. And we look in 3 key areas.

We look at how do we go meet the performance needs at the scale -- we look at how do we extend the flexibility of this core platform to be able to scale not just at the enterprise level from terabytes to petabytes but now into exabytes and tens of exabytes. We extend that predictable linear performance scaling to the needs of the neo clouds. And then we lean on our track record of mission-critical reliability and ability to deliver predictable performance without bottlenecks across the board.

So one of the key aspects here of what we've done with Exa is we've taken that -- the strong routes that we have in FlashBlade Ask that we've proven out in the enterprise -- and we've disaggregated that architecture. We've kind of taken that core product, and we said, "Hey, you're really -- we've built a software base that gives us a very solid foundation to provide one part of the performance puzzle. The metadata piece. We're going to go decouple that software. We're going to go parent with a more open hardware ecosystem and now allow us to provide the utmost performance in both metadata and data because this is the secret.

This is the key to meeting that wide dynamic range of Neo Cloud needs. So we've talked around the term metadata performance and debt performance a lot. Perhaps it's useful to draw a quick analogy, right? If I went and got a book and hand it to one of you and ask you to read a page, right, it might take you a minute, 2 minutes to read the page. If I hand it to a speed reader, somebody who can process data really quickly, might take them 20 seconds. Okay. That's a lot faster. That's great.

Now if instead, I pointed both of you at a shelf of books and I said, go pick the third book off the top shelf, go to Page 17, read 2 sentences and then go pick a book off of the fourth shelf, Turn to Page 237, read a couple of sentences, so on and so on and so forth. You can see that most of the time you would be spending is actually looking up the words to read, not reading the words. And so when you look at the performance required for AI workloads, you need that balance that mix, right, of being able to process the data really quickly, read the words off the page, but more so to be able to process that metadata quickly, be able to find what data is relevant to then go process it.

And so we're uniquely positioned to bring performance to both of those -- both types of performance and to be able to scale both ends of that performance independently and predictably. The other thing that sets us apart is we're able to do this without putting a bunch of onerous requirements on the user, right? If you look at other approaches that try to do something similar in the industry, they come with fairly onerous requirements like, okay, well, we can maybe give you better performance, but you just have to install this little piece of software on the GPU server.

Okay, it doesn't sound too bad. If you're running 1, 10, 15 GPU servers, that's not too bad. Well, if you run in 10,000, right? It sounds pretty terrible. So we're able to build on open standards, extend our core technology, which we've perfected in Purity since day 1 to provide this unique mix and balance of metadata and data performance, exactly what is needed to meet these workloads we're able to deliver it on open standards without a ton of operational headache and complexity to the end users. And that's really what sets us apart.

We'll dive 1 click deeper into this since Paul always likes to remind me that our differentiation is in the software. I want to hammer this home. What is it -- so how do we solve that problem, right? I kind of highlighted how metadata performance is really important. Data performance, I think people understand. So why is this a hard problem Well, it comes down to the fact that organizing the metadata, organizing, when you build storage software organizing the catalog of what files where and where on the media should we go to read this data essentially in my analogy, what -- how do we look up the pages to go read -- this is a hard problem to begin with. It's a hard problem to scale as you scale capacity and keep performance.

The organizational methods for doing this is, frankly, in the industry are quite antiquated. This is a hard computer science problem. The news is we have building all of people that love solving hard computer science problems. And what we've done is we've built the core of purity since day 1 on what we call a key value store, key value database. This is essentially the technique introduced largely by the cloud providers to scale without limit and with extreme predictability and with extreme parallelism. So we built the core of our Purity operating system, whether it's in FlashArray, whether it's in FlashBlade, whether it's in FlashBlade Exa, based on this key value store, plus a set of techniques, a set of set of approaches, I would say that we've adapted from the distributed database community.

If you net all this together, right, if you net all this together, and you look at the software that underlies our products or platforms, whether it's FlashBlade, FlashArray, FlashBlade, what you'll find is a highly performant, highly scalable without bottlenecks architecture that is able to scale very, very predictably and is free of bottlenecks that has served us well in the past. But as we look forward, and particularly in the neo clouds, at the changing workload demands that we're seeing positions us very well and positions our customers very well to be free of changes in the future.

All right. So we start with a strong software base, and then we pair that with unique hardware to bring out the best attributes of that software, in the enterprise with FlashBlade. We paired that purity operating system with a fully vertically integrated hardware environment that accelerates that metadata engine accelerates that data engine, while keeping enterprise-level simplicity, integrated networking, kind of the perfect package to meet the enterprise needs of size, performance, scale, manageability and networking complexity.

As we move to the neo clouds, as we scale up 10x, 100x, right, we expand this platform from a hardware perspective, same software, right? We decoupled the software, but we now pair our FlashBlade hardware, which accelerates our metadata here with purpose designed hardware to serve our data tier right? Basically opening the pipes, allowing for much, much faster data transfer, and much 2 axes of scale, the ability to scale performance and metadata layer, the ability to scale the performance of the data layer and the ability to scale capacity, almost without bounds.

So how does this add up? What does this end up delivering. Well, to start with, #1 in performance, right? We topped the performance charts across the board. As we look at both the top line numbers as we look at the industry benchmarks, which I'll dive into in a second, we're able to serve Neo cloud needs at massive scale, right, whether we're talking about thousands, tens of thousands, up to hundreds of thousands of GPUs. And we're able to do this on the same technology that, again, within the enterprise, we see starting at tens, hundreds and as we scale into the neoclouds, up to 3 orders of magnitude.

And at the end of the day, when you net this out, I started this conversation by talking about GPU utilization when you net this out with a higher performance, with a greater reliability, with the ease of operation, the fact that operators aren't having to go around configuring custom software and spending a ton of time managing the storage layer, we're delivering a significant reduction in TCO for not just our enterprise customers, but now our Neo Cloud customers as well.

I mentioned #1 in performance. MLPerf is 1 of the leading benchmarks in the space. These are recently published results in the newest round of MLPerf 3.0 benchmarks. Key things to call out here. This is the standardized testing configuration. #1 across the board in all configurations tested by a wide margin. The previous version of MLPerf, we're about 2x the nearest competitor. And what I can say is that in this most recent version, some of our competitors declasse -- so that should tell you something there.

All right. So let's take a look at a specific example from a neo cloud deployment. So STN, a U.S.-based Neo Cloud is 1 of our first customers for a when they started their issue, where they started off to the left was they had obviously a large environment of GPU clusters. They were having to serve multiple AI services on top of them. So they're essentially a GPU cloud. Their customers are AI service clouds that have their own service offerings. -- doing a mix of training, doing a mix of inference, lots of data, they couldn't keep up with the performance. They had large multi-petabyte file object stores couldn't keep up with the performance, specifically on metadata.

So what do they do? They went and found a third-party product, separated the metadata, ran a separate cluster to go manage that metadata. So now they've got 1 thing doing the metadata, another thing doing the data, okay, still not fast enough. They put in another product, a cashing layer, more hardware spend, more complexity to sit in the middle, just to provide good enough performance. And at the end of the day, they were left with not that great performance a really complex stack, third-party dependencies and way more engineering effort and time being spent on this than they wanted to do.

They brought in FlashBlade Exa. -- they actually tested our solution in our labs on a Thursday, place the order on a Friday. We had them up and running on a Monday night. They were able to simplify this dramatically, take out all of these extraneous layers, improve their performance dramatically, 3x and beyond, right? So going from being topped out at 30 to 40 gigabytes a second 100 gigabytes a second per node with predictable scaling and they've now grown that environment in multiple environments, multiple fold.

At the end of the day, right, when you ask the CEO of STN and he's spoken on our behalf before, what he said is significant acceleration in storage performance, but what matters the most is the acceleration of the end-to-end inference and training workloads 10% to 20% faster end to end, that includes the GPU time and most importantly, a big uptick in GPU utilization.

All right. So with that, I'm going to bring the morning's tech sessions to a wrap. I think, hopefully, we've given you a sense across the board of all of the things that we're doing with the new product areas to meet the needs of AI, whether it's driven by a new approach to managing and looking at classifying understanding data, how we're bringing that together on a unified platform simplifying the management of that with autonomous policies and then accelerating AI use cases in the enterprise and then now taking this to the Neo clouds.

So with that, we're going to transition to Q&A. I'm going to invite Chad and Prakash and Ashish back onto the stage. And as they're setting up, I'll just remind everyone that we're going to have an afternoon session going deeper into the company strategy and the financial outlook. This morning's Q&A really should be -- is technology and product focused. And then also I would ask just for the folks online that as we the mics around the room. Before the question, if you could just introduce yourself and your firm, that would be helpful. Jason.

Jason Ader with William Blair. I got a couple of questions. The first is just how do you think about the point solutions out there for like DSP data discovery and classification, governance, are these guys competitors or partners? And then secondly, on the NeoCoside, can you just talk about who you're competing in those deals with ?

Yes, I'll start. So yes, One Touch traditionally was working in the DSPM space as a DSP tool, yes, it competes with other DSPM providers. the major differentiators in that category still remain the broadest coverage, like covers everything from mainframe to SaaS tools, et cetera. And a lot of the traditional DSPM tools were built with primarily security-minded Rex classifiers, this is a pattern match. But it could not tell historically whether -- and we've tested that when we were actually looking at One Touch, we've tested a lot of products as we were evaluating 1 to 3, 4, is it an order number or a new invoice number.

Most things would fail on that test, right? So the context in terms of the approach was more built around context and Knowledge Graph where security was a use case and the kind of technology that we looked at. So we saw the applicability to broader than just the SBM category.

I'll let Ashish to add.

I'll just add a little bit to it. We've got an architectural advantage when it comes to looking at data. And it starts with the fact that we don't have to sit in the workflow to understand what the data is doing. We go to the data as a source and create those knowledge graphs and their relationships which allows us to get to deployment a lot faster, glasses to get much higher accuracy is actually much more scalable because you are not waiting for the workflow to happen, you're actually going right to the data and delivering the intelligence right there. So in addition to more coverage, much better context, we also have an architectural advantage to deploy at scale and deliver accuracy.

Can you replace all this...

Well, the whole idea behind the universal data intelligence layer is to provide this intelligence as a LEGO plate on which we can put the various different LEGO blocks -- the SPM focuses on security, but the same intelligence allows you to do things on governance, compliance and then also the contextualization enables you to do AI ready data as well.

Yes, 1 example you would see is most of those solutions is like here's an audit report for DSPMlike yes, this product can produce. But it also provides an MCP for a query what data should be able to go into this agent model. So the products that were traditionally security minded in the SPM focused were built for very narrow use cases, and this is more broad in terms of all entity types, not just security use cases.

And then Jason, to take the second part of your question on the neo clouds. What I can say is we typically don't in the neoClouds, compete with our typical enterprise competitors. We see some folks that have grown up, I would say, out of the traditional HPC space based on technologies like luster and so on and so forth. We see some more niche players that have focused more recently deeply on the neo clouds. What I'd say that sets us apart from these players is neither of them are able to address the wide set of performance needs that I articulated between data and metadata performance, needing both of those at the same time as well.

Increasingly, I would say the focus, as we have conversations with NeoCloud customers is on the gaps and the competition as it pertains to reliability, operational simplicity and availability. Like that is becoming front and center, much less security, by the way. But those aspects are becoming front and center as these neo clouds are, I would say, graduating into truly mission-critical workloads that have compliance that have governance have other requirements around them.

Erik Woodring, Morgan Stanley. Rob, I appreciate the color that you shared on FlashBlade exit there and maybe building off of that last question. I'd love to know like do you think has to change for evercore to become like, let's call it the default storage platform for the next 10 to 20 major Neocloud build-outs? -- what changes for you to become the de facto rather than trying to get into them to speak? .

Yes. Erik, thanks for the question. I mean I think 1 thing is to be perfectly clear, we were not first to enter this market, right? We have focused on rise we, I would say, made a very intentional and concerted effort to adapt our technology for where we thought the needs of this market would head. So we're not first entrance. So in many cases, it's about going and engaging and we are engaged with pretty much all the top Neo clouds. But as you might imagine, these are longer sales cycles, very technology-focused. .

But I think it's a matter of time seasoning and realization that these environments are super high-performance demanding across the multitude of dimensions and they have all the mission criticality requirements, and we're the only vendor that can meet those needs. The other thing, I would say that a cruise store advantage and what we're seeing really good signs of already is People, again, forget that Neo clouds are clouds. Clouds have other storage needs beyond high-performance GPUs. They have things like IT departments.

They have things like virtualization needs for the user sessions. They have file storage needs, object storage needs. They even have backup needs. So our ability as a holistic vendor to go and meet the needs of these new clouds, whatever door we can get into allows us to penetrate and accelerate the conversations around FlashBlade Exa and the high-performance storage in ways that nobody else can.

Mehdi Hosseini, Susquehanna International. It's a question for the team. When I look at your slides and think about application, it's mostly a GPU-based accelerator open source, which makes sense given your product portfolio as you look into the future, what happens when we actually have more of a custom ASIC solution, which seems to me would require more proprietary operating system. So the question is, am I right to assume that right now, your data management solution is mostly applicable to open-source GPU base? And if that's the case, how do you extend that to more of a custom ASIC proprietary operating system.

Yes. Maybe I'll start with that one, and others can jump in. Thanks for the question. I think what's important to realize is everything we've talked to you about this morning really is driven by software, whether it's Data intelligence, whether it's what we're doing with data stream and the organization of data for AI and even go what powers EXA is software-driven. We've taken a very intentional approach to design that software, not to be specifically locked into any 1 execution platform, whether it's specific CPUs, GPUs or accelerators, we obviously keep abreast of new hardware developments in the market.

But as we -- as it pertains to some of the areas you mentioned, whether it's data intelligence, data stream, we are designing those layers and those modules, the LEGO blocks, as Sesh mentioned, in a very modular way to be able to run on any hardware platform. They'll run great on CPUs, they'll run great on GPUs. If some Neas comes along, that makes it run even better, great, we'll run on that. data stream the same way.

We developed that in partnership with NVIDIA on their AI data platform. We're working with a working with other GPU vendors to run that same technology. It doesn't -- it's not in our benefit to lock our software to any 1 hardware platform. We're going to make that work on the best hardware platforms that exist today and what we see coming in the future.

I just think about your unified data lake, unified data plane -- it sits either below or on top of the operating system. Whether be open source or more proprietary. As I understand that you don't want to be specific or you don't -- you want to be independent of silicon, but what about the operating system. How is the unified data plan would work with operating system.

I can jump in on that 1 quickly. So if you think about what the UNFI data plane does effectively is provides open protocols for storage, right? So we do block file as well as object in a 1 common mechanism. No matter what the hardware is on the other end of it. They are receiving that type of protocol stack. And even within block, we have various ones of fiber channel and NVMe over fabrics and over TCP as well as other ones that are out there.

So I'd say we follow pretty much the networking connection as well as the things that operating systems will consume storage around and provide that as one unified view. And I think 1 of the benefits here is that we have products that actually cover multiple protocols. FlashBlade, as an example, does file an object to where FlashArray does block file and object together. The control plane starts to become the management component treated side where it actually lands based upon the SLAs and intent that user defines.

And so it rolls through the operating environment no matter what the hardware system looks like.

So Steve Cabot, Morgan Stanley. So Paul keeps telling us that you're a software company, and there are a lot of other software companies out there and -- many of them also are very interested in noting the data governance stacking, et cetera. It's sort of the right to contest the market, but there's also what is the technology dividing line where you go up the stack, they're people trying to come down Snowflake, Databricks, Okta, all the security guys.

So who does authentication, who does -- where do these policies come from? And I guess I'm trying to -- we can either construct a framing where it's a battle royal where you guys are slugging it out there or you guys can find some kind of a clean place where you do this, I do that and everybody is happy. How does this all evolve? I agree it's going to evolve. The key question is what's your right to contest, where do you define a line? And what is your relationship to those guys?

Yes, I'll start and then you can win. So largely, I think -- and I talked a little bit about it earlier. If you take A lot of those vendors are taking a -- like come from a different point of leverage, there a point of leverage that Oracle comes from the data, it was a base of leverage. Data Brooks comes from the analytics point of leverage sales force at SAP even come from an application point of leverage. There's -- so the data management ecosystem has been brought and grown up traditionally for a really long time. Most of them have largely not come from solving the problem needed for AI today.

So all of them are trying to build capabilities bring your data to me, and I will do an ontology and knowledge graph on things in my ecosystem, right? So just 1 philosophically, we're approaching the problem very differently. We're saying, no, your data is not going to go into 1 of those ecosystems. It will be all of them. It will be Databricks and Snowflake and SAP and Oracle. And your data will go and to be in multiple places. That's the this concept of data will be distributed. So oncology and knowledge graph needs to be in middleware above the layer.

And I'll use an example of Microsoft has capability in their platform with purview. They tried to do it for Microsoft data. And some we've got customers that are doing that, but then also use our data intelligence technology because Microsoft can't look outside. So this idea of if you believe data is distributed and you need to work across multiple ecosystems, we provide a neutral way to work across data wherever it sits. We're not opinionated on it needs to be in Databricks or Snowflake or anybody. And all of the other vendors have an opinionated view of if you want any value first, give me all your data, right? So philosophically, we found that, that resonates with customers very differentiated from the market.

Now in terms of are we slogging it out, are we partnering? We're about a lot of conversations with and some of those vendors saying, they're interested in partnering with us for the things that they can't see or do, right? So there will be -- there might be co-opetition in the direct hey, everything needs to be here. But as they start accepting that they don't have a ownership of all of a customer's data than they need this external view and we provide that external view into a customer's landscape.

From a customer lens, we're just -- we mainly stay focused on customer value. We've always done that from the beginning. And the customer is being total visibility into their data wherever it sits. They have multiple vendors. We see that in almost 100% of our installed base. And we provide that so they can remain in control and get value of the data where it sits. So that is largely how we're approaching it, but feel free to...

I mean I think just to add on to that, 2 things. One is, I think the 2 up slide that Prakash put up puts in very well, which is we're pursuing an approach that adds value -- seeks to add value on top of those platforms. And that value largely is in building the connectivity, the context, the universal context that allows customers to more usefully connect it across those sources, whether it's SaaS vendors or the likes of Databricks and Snowflake.

Another example, we have a partnership with Databricks today, right? And that partnership is allowing customers to connect the Databricks processing engine to Flashblade to data stores start and FlashBlade on-premise. And I think it's a good example of an early example of Prakash is talking about, that partnership is driven by the fact that customers have these data lakes, they for some applications benefit from Databricks capabilities for other applications, that source of data needs to be connected to other parts of their enterprise, so they need to control that thing and they want the flexibility to be able to burst and say, hey, for this particular task or analysis, I want to use a databricks engine.

For this other set of needs, I want to use a different application environment. And so as we zoom back out and we think about our vision of data primacy and how this plays out in the future, I think it's a great example of a first of many steps that are going to take place, which is to say, the true store of value is in that centralized data thing. And then the application or platform stacks over time, evolve to who can do a better analysis job, who can do a better process encoding for financials stack or an ERP stack separate from the data.

And also be able to optimize like the workflows to it, right? So there's a lot of benefits for us to build good partnerships and efficiencies in the way those processes access our data very similar way that we did like early days with VMware for virtualized environments, we made it very easy for customers to be able to embrace the technology. We're going to do very similar mechanisms there as well.

Howard Ma with Guggenheim Securities. I want to ask a higher-level question, but I also want to tie it into the prior question. which is -- so first, the high level one, which is when you talk to CIOs, CTOs and CISOs, I imagine like when you offer this unified approach, a Discover data wherever it suits provide the context and the governance, you get very little pushback to that concept, right? But I'm sure you get inertia because there's a lot of existing source systems in place already.

So question number 1 is you're in 2,300 customers today with Fusion/EDC. Can you give us a sense of how those combos have evolved? And then the tie-in to the prior question is, as they consider ever beer with the existing tech stack. I kind of want to summarize it. It sounds to me like, obviously, the database layer still needs to stay right, whether it's OLTP analytical, a some parts of security like I authentication it seems like those are crucial, complementary. But it sounds like you're saying data integration, like MDM, DSPM, maybe even inference gateways, like there are categories that if you sum it up, it's probably like tens of billions of dollars of TAM. That is what you're directly going after and that's what these C-level people are considering replacing with Everpure.

Yes. Maybe I'll start this kind of long compound question. So a couple of things. Absolutely, as we talk to CIOs, CDOs and I would say, decision-makers hypertenstack, not only are we not getting pushed back, we're actually getting significant pull I've been on the road as have the rest of the management team since the acquisition.

And I will say every single meeting that we have with a high-level customer prospect -- the first topic and really the thirst in demand is for a lot of these capabilities. One thing I would correct maybe or modifying the question is important to realize that as we deliver One Touch, as we deliver on that road map into data intelligence, these are a set of capabilities we can offer on top of anybody's storage, right? So the existence of other storage infrastructure in place does not get in our way.

Yes, the benefits of having 2,300 customers on fusion and growing, the benefits of having 15,000 customers on our core platform is -- will accelerate that. But this set of buyers is very much -- it's very much needing and wanting and thirsting for better control understanding of their data, where it sits for all the reasons that Prakash and Ashish outlined. And then I think to the second part of your question in terms of, hey, as we look forward and we look at the traditional markets, if you will, of data management, whether it's MDM, cataloging so on and so forth. What I'd say is that we -- in many ways, just like we've done with storage infrastructure, we're approaching the market needs with a little bit of a different lens.

We're not just seeking to reinvent and or replace existing markets. We see an opportunity with technology from One Touch with new technologies coming out of AI. We see an opportunity to fundamentally disrupt and do that thing better, right? And so if you think about a lot of why -- and Prakash set us up well in the beginning of his discussion, if you think about a lot of why the enterprise architecture problem has been difficult in the past, it's because of the seams between all of those pieces, the MDM, the data catalog and the integration and the ETLs, and a lot of that can go away.

Our mission is to make most of that go away and to go away in service of separating out the data layer, having that become primary and then relegating the application stack to becoming more stateless encoding of business process over time.

Yes. And look, just to build on it, and I'm going to go back and probably date myself, but I was around when at SAP, we had something called R2, which was more mainframe less and not client server, right? So -- and that was there when we were moving to client server right? So -- and I was there when web and Internet technology kind of came about. And we've had platform modifications and cycles of how we build applications over time. And I actually do think this is a long-term fundamental shift in terms of the way you build applications, like in the same way, we now assume client server or web micro services.

I do think a data primacy architecture will be a 10- to 20-year super cycle. Like I do think that is a new way of thinking. I do think AI has forced it. I think, has created an inception point for us to think about it differently. As Rob said, we're not chasing things, and we know things don't change overnight. But today, when we build an application, and we think about how customers are going to build applications, right? The idea that I'm going to buy a ton of applications and integrate to the data versus AI needs markdown files and context and includes source and config and logic and pricing rules and like business logic as data like the business logic as data is a superior approach for building modern applications.

And this might be just a long-term frustration of mind saying, we never really got the ERP because the application world became fragmented. I actually believe if the ability to build application gets shrunk down by AI development tools because software engineering is 1 of the things AI is getting good at then the cost to build workflows comes down and the cost to maintain data coherence goes up. So by starting with the data primacy and data intelligence right now, you can create a baseline of data coherence.

And I do think that is a way like new startups today will build applications that way, right? New companies will. The long enterprise and everything that we have today won't change overnight. But the more the simpler the app, I would say there's a category of applications, and I won't say who these companies are basically a web form on a table. I don't know that you need to buy those apps. The category applications that have deep embedded process logic, probably not the first thing that's going to change, right?

So I do think that as this thing evolves, data intelligence is the first step looking at the data where it sits, understanding the relationships and ensuring that both analytics apps and agents can work off that data with shared context is a first step. And that's why we're taking this step. It's not the last step. Data intelligence is our first step in this journey.

Asiya from Citi. Just a quick question on the adoption that you talked about, I think you said 23 customers, 2x adoption in the last -- is in the last 6 months. Are these existing logos that you're already present at -- are you seeing this? And I'm sorry if I missed this, but -- or are these new customers that are approaching you given the problems that you guys have talked about? And so opportunity should be pretty huge if it's not already in the installed base?

Yes. 2,300 customers, 23 would be a little bit small. So it's both, right? So existing customers obviously see significant value out of it because they're helping to build intelligence at top of the existing infrastructure. And I'd say the benefits on the existing customer is they're starting to see this experience shift and looking at their other environment saying, why do I continue to do what I do with those other older legacy products. And so you're seeing kind of a push towards I want to move to that new experience and a consolidation around this kind of unified data plan model.

On the new customers, we've been talking to them about this experience kind of from the get-go. And so we start them off fresh in that minute, -- and the game here is that you can have conversations that orient them in a very different direction. You're starting to talk to them about business processes and requirements and what intent that they have as a, call it, first principle to start with. On the existing customers, we actually have the benefit of telemetry. And we have lots of data that comes back from each of these systems and so one of the first things we're doing is helping them rationalize what do you do today?

And what's crazy is most people say, "Hey, we only have 3 policies for this, and we can show them they have like 55 policies. And that's the problem of lack of consistency. And so the first benefit we get is actually being able to help them align business requirements and policies and then apply that to the infrastructure stack. And the nice part here is when they look at the other environment that they have, they realize, "Oh my God, I can just imagine how much of a mess that is. I would really like to get more consistency and governance built into the system and hence as those things come up for refreshes Lots of it is landing back on this new experience. So both existing and.

One of the things I think I shared with you this morning when we were speaking just over coffee was this idea that Pure has consistently taken share in market over time. I think Charlie may talk about that like a little later today. But as we take a look, some of the main drivers of that, our SLA approach that Chad talked about, is something that is bringing more new net new logos given the unique differentiator we have to the competitive market space. The data intelligence approach also, we see signs of opening net new logos for our kind of growth business. So we've always done well in taking share as a baseline, but kind of those areas of the business have a higher net new logo concentration of kind of a front door because it broadens our value proposition into the market.

This is MP on behalf of Jucarduso from JPMorgan. Just a quick question. You talked about different architectures across enterprise solutions as well as the new cloud. any differentiation? Or what's really different in terms of offerings or overall solution for hyperscale customers?.

Yes, I'll take that one. Yes, and we'll dive a bit deeper into this in the afternoon session. But yes, I would say if we look at the different levels of scale and customer segments between our typical commercial enterprise customers and their needs, both of performance scale, simplicity size, et cetera. what we're seeing in the neo clouds and then now the hyperscalers. -- same technology that underpins all of those solutions in terms of the software base, in terms of -- but in terms of the configuration of the offerings, we've specialized those over time, really for those 3 segments. I'll dive deeper into the hyperscale solutions in the afternoon session. .

But really, the takeaway point is it's all based on the same core underlying technology and really packaging and integrating that into each of those environments, which have very different needs, right? If you look at the enterprise, they typically want a 1 size fits all, put it in the corner, plug it in and the thing just works type of solution as you go 100x in scale to the neo clouds, -- they're now delivering a cloud service. They have SREs, they have operations teams. They're operating at a much larger scale, have different set of requirements in terms of performance, efficiency, how they go and manage systems.

And then if you upsize that again by 100 or 1,000x, now you're talking about hyperscalers, as you might imagine, each of those orders of magnitude carries with it a very different environment and different operating requirements.

Matt Katri from Needham. When you think about some of the earlier adopters of Fusion and even EDC, like before you introduced this unified data plan, is data intelligence something that they were building out themselves? Or is it just that new of an issue? And like kind of going back to like why is Everpure at the right layer to approach this from? When you talked about it's a fundamentally different approach from like a Snowflake or data bricks. How does that change now that they've dove into like interoperability and iceberg tables and things like that.

Yes. Okay, I'll start. So kind of 2 parts to that question, right? The first part of was data intelligence typically needed in the past in building EDCs. And typically, from a traditional storage lens, I would say no. Prior to AI, storage team who managed kind of north-south applications. This is my Oracle this is my SAP. This is -- and it was sufficient. -- right? Our enterprise data cloud was built around intent and policy management and the 2,300 customers that Chad talked about were all kind of pre the data intelligence era. Why data intelligence became important is when previously, the predominant paradigm was application centricity.

Everything was by an app, deploying app, right? That was most of what was going on. When agents start working on data directly, the paradigm shifted. So now you need to look at and manage and understand that data. You need to contextualize that data, right? And from a storage management lens, that means, like I said earlier, if you have an agent and you need high availability for it. So it can be resilient that is a storage problem. -- and it is uniquely a store problem, right, in terms of how you can offer that capability.

So doing the data management, like of that right thing, requires bringing that data context closer to storage. That is the leverage point that is the uniqueness we see in terms of why A+B is kind of uniquely differentiated. Coming back to your kind of -- so like why data intelligence and why now AI. It was just a change in inflection point, right? And that happened faster probably than I think most of us would have expected.

The second part of that discussion is that data intelligence advantage around why at this layer, why storage, et cetera, there's a more near-term and a longer-term answer. -- kind of in the more near term, you just want to ensure that hey, anything that happens physically doesn't break context upstream, right? That's a near-term storage problem. -- where storage steering, storage mobility, storage things, retain context and memory.

Otherwise, you're AI spending tokens, retraining and learning on that data, right? Longer term, that leads to more persistent memory and a storage operating system. So if you do this at higher layers in the stack, right, you then have different resiliency challenges, different availability challenges and you end up back in that wild Garden strategy I was talking about earlier, saying, okay, well, that just means putting all applications like every application provider has to build that uniquely and you fragment the management of that thing.

And maybe just to zoom out and kind of put all of that in context because we've turned a lot of you guys this morning. Maybe the way I'd think about it in more simplistic terms is the unified data plan, that's been our mission since day 1, build the best storage infrastructure, do it 1 way, do it across all of the different storage needs that an enterprise has and make that a unified experience and very seamless across the board.

When we think about the intelligent control plane and management, that's about saying, hey, once we've done that first thing, -- how do we let customers not manage 100 discrete arrays, but manage that as 1 pool of resources like a cloud of storage, if you will. And then that then puts us in a unique position to say, okay, well, what more can we do now that we are helping customers with the best rage infrastructure, managing it well, tying into policies -- that now puts us in a position to go add value at the data meeting, data intelligence layer, both on our equipment and arrays as well as other environments.

So with that, kind of a couple of minutes over time. I think this brings the morning session to a close. Wanted to thank all of you for the great questions. I think we're going to adjourn next door for lunch and be back at 12, is that right?

Thank you.

Good day, and welcome to the Everpure Second Quarter Fiscal 2027 Financial Results Conference Call. Today's conference is being recorded. [Operator Instructions]

At this time, I'd like to turn the call over to Paul Ziots, Vice President of Investor Relations. Please go ahead.

Thank you. Good afternoon, everyone, and welcome to Everpure Second Quarter Fiscal Year 2027 Earnings Conference Call. On the call, we have Charlie Giancarlo, Chief Executive Officer; Tarek Robbiati, Chief Financial Officer; and Rob Lee, Chief Technology and Growth Officer. Following Charlie's and Tarek's prepared remarks, we will take questions.

Our press release was issued after close of market and is posted on our website where this call is being simultaneously webcast. The slides that accompany this webcast can be downloaded at investor.everpuredata.com. On this call today, we will make forward-looking statements, which are subject to various risks and uncertainties. These include statements regarding our financial outlook and operations, our strategy, technology and its advantages our current and new product offerings, our ability to procure sufficient supply of components and manage our supply chain, our hyperscale opportunity and competitive industry and economic trends.

Any forward-looking statements that we make are based on facts and assumptions as of today, and we undertake no obligation to update them. Our actual results may differ materially from the results forecasted, and reported results should not be considered as an indication of future performance. Discussion of some of the risks and uncertainties related to our business is contained in our filings with the SEC, and we refer you to these public filings.

During this call, all financial metrics and associated growth rates are non-GAAP measures other than revenue, remaining performance obligations or RPO and cash and investments. Reconciliations to the most direct comparable GAAP measures are provided in our earnings press release and slides. This call is being broadcast live on the Everpure Investor Relations website and is being recorded for playback purposes. An archive of the webcast will be available on the IR website and is the property of Everpure. Our third quarter fiscal 2027 quiet period begins at the close of business, Friday, October 16, 2026.

with that, I'll turn it over to Charlie.

Thank you, Paul. Good afternoon, everyone, and welcome to Everpure's Q2 Fiscal 2027 Earnings Call. Q2 was another outstanding and remarkable quarter. We exceeded our guidance range and all key company metrics. Revenue growth of 38% year-on-year continued our 30-plus percent growth performance since Q4. Operating profit surged 77% to $230 million year-on-year. Growth was broad-based across all geographies, products and business segments. Strong sales momentum carried over from Q1 and continued to build throughout the quarter.

Additionally, TCV for Evergreen//One has accelerated to a $1 billion run rate for fiscal year '27, which indicates that our market share gain momentum is even greater than our revenue growth implies. We have seen our revenue growth accelerate, consistently and steadily, over the last 8 quarters, and we now believe that this higher growth rate will be sustainable for some time. Based on current demand signals and win rates, we believe we will see sales and market share strength continue into next year, as you will see in our revised full year guidance.

Only some of our growth acceleration can be attributed to price increases as our growth is far beyond that of our legacy competitors. We believe that we have entered into breakout territory in our core enterprise market, because of the steady progress we have made in building out our product line and architecture. I'm going to speak to you today about 4 areas that underpin our confidence in our long-term growth.

First, the cause and foundation for the acceleration of our growth and market share in our core market. Second, how the current pricing environment advantages our technology and company. Third, our growth opportunity in new areas of business, such as AI, data management and modern apps and virtualization. And fourth, our growing hyperscale opportunity. First, let's look at what's driving our growth. Everpure has spent the last decade expanding beyond our groundbreaking flash array product to serve every storage segment from backup to AI, from TeraBytes to ExaBytes, all on a single unified software foundation, Purity.

Purity unifies block, file and object. It drives DirectFlash and allows our customers to benefit from products that never grow old with its unique Evergreen capability that promises nondisruptive upgrades forever. For example, one of Europe's largest telecom operators, a long-standing customer enjoys flexible scaling without disruptive hardware refreshes, significantly lowering their total cost of ownership.

During the last decade, we also invested in the ability to have our capabilities offered as a service with Evergreen//One and offered on the cloud with Everpure Cloud Storage. As a senior executive from a fast-growing global managed service provider stated, we chose Evergreen//One because it gives us predictable, consumption-based economics and lets us expand capacity and performance as customers' demands change. With Everpure managing the underlying infrastructure and life cycle, our teams can spend less time maintaining storage and more time delivering reliable, high-value services to our customers.

That's why they have fully embraced our SLA-backed Evergreen//One model. We alone allow our customers to operate our systems as their own global enterprise data cloud with Everpure Fusion. And now we are enabling our customers to contextualize their data to make it AI-ready with Data Intelligence from our 1Touch acquisition. Since our very first product, industry analysts have recognized Everpure as an innovation and customer leader. Just last week, Gartner validated that momentum once again positioning Everpure highest in execution and furthest in vision in their 2026 Magic Quadrant for Enterprise Storage Platforms.

Alongside that innovation, we have scaled and developed our go-to-market engine and strategy to become a top-tier competitor across mid-market, enterprise and government markets globally. Our brand reputation has expanded to where every customer needs to consider Everpure as a supplier in every area of data storage. It is the combination of all of these factors that has brought us to this point, where our growth has reaccelerated and it gives us the confidence that this higher growth rate will endure for some time.

Next, let's address the current pricing environment and why it directly advantages Everpure. The rapid increase in semiconductor demand and cost continues to affect our industry in multiple and complex ways. This quarter, we saw the predicted effect of increasing prices on both sales and demand. Customers transacting now are paying more for less capacity, while others are increasingly adopting our as-a-service model. For example, this quarter, the University of Western Australia expanded its Evergreen//One footprint with us, enjoying the ability to scale its storage on demand while managing costs efficiently.

Evergreen//One has grown substantially in this environment and is now on a total contract value or TCV run rate in excess of $1 billion this year. Everpure's advantages in flash technology, lower operational labor costs and leadership in our Storage-as-a-Service offerings are providing us with outsized market share gains in this high-cost environment. And our decision to honor our past and existing commitments, share the burden with our customers and operate at the lower end of our product gross margin range, while components costs escalate, has solidified our relationships with both customers and channel partners.

Adjusting for the acceleration of Evergreen//One, which dilutes reported near-term revenue growth numbers, we are growing well above 40% year-over-year, far exceeding the market and our data storage competitors. Beyond our core portfolio, we're opening up major new avenues for growth and expanding our market opportunity. The growth of AI is focusing customers more than ever on how they manage their data. At our June//Accelerate user conference, we introduced the concept of Data Primacy which will be the future of IT architectures in the AI era.

Data Primacy posits that for organizations to streamline their operations and make their data AI-ready, they need to rationalize and structure their data into sources of truth and systems of record. Data Primacy will enable enterprises to take control of their own data, manage it strategically in their own data cloud, and provide governed access to their data and its context to their chosen AI service providers. Interest in Everpure Data Intelligence, which includes capabilities from our 1Touch acquisition is off the charts. Every customer we meet identifies with the core issue facing them in their use of AI namely that their internal data is far too fragmented to be immediately useful for the AI future.

Fragmented and inconsistent data leads to lots of manual reconciliation and incorrect results. While a 95% accurate answer to an AI search prompt can be considered a great result, a 95% accurate invoice is unacceptable. Businesses require 100% accuracy in their financial and business operations. 100% accuracy depends on data that is unambiguous. Everpure Data Intelligence allows enterprises to find their distributed sources of data, discovered their semantics and context and deliver a shared context between the different data sets. It can find both the similarities and the discrepancies between individual data elements and make this available explicitly to the organization.

Most importantly, it allows the enterprise to control these capabilities themselves, rather than trusting their data and context to third parties while enabling selective access to their AI service partners. This puts the enterprise in control and gives them full sovereignty over their own data. It is the first step to Data Primacy, and it is an exciting new area of growth for Everpure. Our Enterprise Data Cloud strategy continues to advance with over 2,000 of our 15,000 customers now enabled with the capability. Customers appreciate the ability to set policy for their global storage fleet and have the systems automatically and reliably configured to comply with their company policies without error. Our Pure 1 copilot lets customers monitor, diagnose issues and deploy changes fleet-wide, making global data management easier, more consistent and more secure.

We continue to scale sales of our products supporting the AI needs of both Enterprise and large-scale AI neoclouds. This quarter, Samsung Electronics chose FlashBlade//S to support its advanced AI and research initiatives. And a major global bank selected FlashBlade//S to standardize their internal GPU-as-a-Service and agent development and deployment environments. Everpure Data Stream, our offering built on NVIDIA's AI data platform was released for general availability and saw its first sales this quarter. One of the largest trial court systems in the U.S. selected Data Stream to streamline and automate AI accelerated analysis of petabytes of unstructured data, saving hundreds of hours of specialized labor and transforming decades of court records into an AI-ready knowledge base while meeting the court's strict data privacy requirements.

FlashBlade//EXA, designed for the highest performance needs of large-scale AI continues to expand. An international high-performance and AI computing provider selected EXA for their highest performance AI training environments. And STN, a large AI managed service provider, expanded its FlashBlade//EXA investment to support growing customer demand. //EXA gives STN the world's highest performance and scalability while simplifying operations. FlashBlade//S and FlashBlade//EXA in combination, uniquely positioned Everpure to meet customers' AI needs from the smallest Enterprise AI environments to the largest neoclouds and everything in between. We've seen a steady increase in GPU attachments for our FlashBlade products across the spectrum.

We see growing traction in our virtualization solutions as customers evolve their VM estates. Everpure continues to help customers optimize their VMware footprint with VCF. While adoption for non-VMware modern virtualization solutions, such as Portworx with Red Hat OpenShift and Nutanix virtualization, continue to grow. Everpure has been named Technology Partner of the Year by both Red Hat and Nutanix, and our VMware alternative solutions have grown to well over a $100 million per year run rate.

As we announced on August 10, we are pleased to have achieved a design win and signed a supply agreement with a second top-5 hyperscaler for our hyperscale products. This newly signed agreement will begin to generate significant revenue in our FY '28 and beyond. As discussed in previous quarters, we expect hyperscale product revenue to rise significantly in Q3, growing in Q4 and in the years ahead. This win confirms DirectFlash's value proposition in massive scale environments with customers that build their own storage services. DirectFlash is software-defined flash, streamlined, efficient, advanced software, providing media management, resiliency, performance, quality and reliability paired with reliable hardware.

Moving the complexity out of black box, SSDs and hard disks lets hyperscalers optimize storage for their workloads, delivering industry-leading density, power, cooling and lifetime. DirectFlash provides customers the flexibility to evolve and qualify faster as their infrastructure scales enables a resilient supply ecosystem across a range of flash technology. With the second top-5 higher scalers signed and other prospects in progress, we will now refer to our sales and strength in this business in terms of our hyperscale solutions as a whole, rather than referring to any specific customer, customer details or prospect time frames.

We expect large hyperscaler order commitments for our DirectFlash solution to extend into calendar 2028 powering tens of ExaBytes of capacity, represent a multiple of our expectations for this year. Our estimates and committed order volumes for total sales to hyperscalers through both fiscal year '27 and '28 continue to grow and will be incorporated and consolidated into our product revenue and guidance reporting.

We look forward to providing more details on our product advancements, particularly our core growth and expansion into new areas, including advanced data management at our upcoming International //Accelerate conferences starting in September, as well as our financial analyst meeting scheduled for September 23, here in Santa Clara. We continue to operate in a very dynamic macro environment and a tight supply market, yet demand for our products is strong, even amid substantial price increases across the industry. This is yet another of many reasons why we feel confident in our growth, our strategy and our future.

I'll now turn the call over to Tarek to provide greater insight into our performance and our expectations for the remainder of the year. Tarek?

Thank you, Charlie. Q2 was another outstanding quarter for Everpure and represents an important milestone in our growth journey. Year-over-year, revenue increased 38%, while operating profit grew 77%, with both metrics exceeding the high end of our guidance range. In addition, this marked our third consecutive quarter with performance above the Rule of 40, underscoring our ability to deliver both strong growth and profitability.

Importantly, our performance was not driven by any single product, customer or geography. Growth was broad-based across our business with strength across all major geographies, product categories and business segments and the sales momentum we experienced in Q1 continued into Q2. As we look ahead, we believe the fundamentals supporting our business are more than ever firmly in place. We are entering Q3 with a healthy pipeline, strong customer engagement and continued momentum across the organization. As a result, we're raising our full year revenue growth guidance substantially to reflect the strength of our execution and the confidence we have in our outlook for the remainder of the fiscal year. More on that, later.

In addition to our excellent financial performance in Q2, on August 10, we announced that we secured a design win and supply agreement with a second top-5 hyperscaler. This achievement represents a significant validation of our technology and strategy. Securing a second top-5 hyperscaler reinforces the compelling economic operational and performance advantages of our DirectFlash technology and demonstrates the growing recognition of our differentiated architecture in the world's most demanding data infrastructure environments. As we noted in our August 10 announcement, we expect only a de minimis revenue contribution in fiscal year '27 from this new agreement and anticipate a meaningful ramp beginning in fiscal year '28, and with revenue continuing to scale in subsequent years.

Importantly, this agreement marks another inflection point for our hyperscale products and provides another long-term growth opportunity for the business. Now let's deep dive into our Q2 performance. The exceptional momentum we experienced in our core in Q2 was driven by a combination of pricing, mix shift and capacity growth, offsetting lower system unit volumes. Our ability to implement price increases while continuing to drive demand, demonstrates the strength of both our technology and our customer relationships. We saw particularly strong demand from large, established enterprise customers, many of whom were willing to absorb higher costs to secure access to our solutions. In fact, we are operating at a price level that we have not seen in 10 years. Even as we tested higher price points, demand remained resilient, particularly among our top-tier enterprise customers reflecting both the mission-critical nature of our solutions and the value customers place on our differentiated offerings.

Product revenue increased 54% year-over-year to $687 million. As expected, hyperscaler revenue contributed only minimally during Q2. It is worth noting a couple of observations that attest to the overall strength of demand for our products. First, customers are buying fewer solutions at higher prices. Second, we have observed a mix shift to higher performance configurations with TeraByte capacity up across our portfolio. Third, in terms of deal sizes, deal above $5 million grew 59% and deals above $20 million grew a whopping 385% year-on-year. Finally, sales of our Evergreen One solution have grown significantly higher than product sales, indicating even stronger growth than appears in our current revenue numbers.

Overall, Q2 demonstrated the strength of our business model, the durability of demand across our customer base and our ability to execute in a dynamic market environment. We remain focused on sustaining this momentum as we continue to expand our market opportunity, deepen customer relationships and drive long-term profitable growth. Turning to our Storage-as-a-Service business. We continue to see exceptional momentum, particularly with Evergreen//One. Evergreen//One total contract value, or TCV, has accelerated to an annualized run rate of above $1 billion for fiscal year '27.

Evergreen//One performance in Q2 was driven by a combination of both velocity deals, those below $5 million and large enterprise agreements, highlighting broad-based customer adoption. The current pricing environment has further strengthened the value proposition of Evergreen//One. Unlike traditional product sales, which can be more directly affected by component cost fluctuations, Evergreen//One is built on long-term customer commitments with lower upfront capital requirements. With Evergreen//One, customers can ramp into growth and are billed on a consumption basis which allows them to better match expense outlays to the growth of their solutions.

These characteristics provide customers with a more predictable and cost-effective operating model. Importantly, and because we control the configurations of the solutions that underpin the Evergreen//One SLA-based contracts, we were able to contain price increases for Evergreen//One, well below the price increases of traditional product purchases, making the offering even more compelling in the current environment. As a result, TCV for our Storage-as-a-Service portfolio, which includes Evergreen One, increased 121% year-over-year to $277 million in Q2. The continued acceleration we are seeing reinforces our belief that customers increasingly value consumption-based infrastructure models that provide greater flexibility and cost predictability.

Turning to our broader subscription business. Subscription services revenue in Q2 increased 20% year-over-year to $499 million and represented 42% of total company revenue. Annual recurring revenue, or ARR, increased 20% year-over-year to more than $2 billion, driven primarily by the continued growth of Evergreen//One and strong renewal activity during the quarter. Remaining Performance Obligations, or RPO, increased 44% year-over-year to more than $4 billion. Growth in RPO was driven by strong bookings across our core business including offerings attached to Evergreen//Forever as well as continued momentum with Evergreen//One.

I'd like to briefly address the relationship between RPO and ARR, as differences in the timing of these metrics can create variations in their respective growth rates from quarter-to-quarter. RPO reflects a total value of contracted revenue that has yet to be recognized, while ARR measures the annualized value of recurring revenue currently contributing to the business. Because multiyear contracts are included in RPO when they are signed, RPO often serves as a leading indicator for future ARR growth. As a result, ARR can temporarily lag RPO during periods of strong bookings activity, particularly when large multiyear contracts are signed. Given the acceleration we are seeing in RPO, we would expect ARR growth to continue over the next several quarters as these contracts begin contributing recurring revenue.

Turning to gross margins. Total gross margin was 69.9%. Product gross margin stood at 66.2% in line with our long-term range of 65% to 70%, and representing an increase of 70 basis points sequentially, while subscription services margin was relatively unchanged at 74.9%. As predicted on our last earnings call, the revenue contribution from our hyperscale business was minimal in Q2, and we continue to expect the majority of hyperscaler revenue to be recognized in the second half of fiscal year '27. Hyperscale product deployments yields margins in the range of 75% to 85% and therefore, should provide an incremental effort to total product gross margins as volumes begin to ramp in the second half of fiscal year '27.

Let me address product revenue growth and gross margins, excluding hyperscaler revenues. First, the pricing actions we implemented in Q2 have largely offset the increases we've seen in component costs so far. We continue to monitor component costs to maintain stable pricing and margins. Second, it's important to emphasize that our pricing strategy remains focused on balancing near-term profitability with our long-term growth and market share objectives. We have approached pricing in a measured and disciplined manner remaining consistent with our commitments to customers while preserving the long-term strength of our franchise. Our goal here is not simply to maximize margins in the current environment. We intend to continue to operate at the low end of our 65% to 70% product revenue gross margin range to drive top line growth and market share gains while maintaining strong customer relationships.

Ultimately, our plan is to allow our product gross margins to return to the upper end of the long-term range of 65% to 70%, once semiconductor costs begin to stabilize and return close to original levels. This balanced approach is clearly paying off as we are driving higher levels of top line growth and operating leverage as our operating profit growth attests. Our operating profit of $230 million grew 77% year-over-year, resulting in an operating margin of 19.4%. Revenue outperformance and pricing management discipline drove this excellent result.

With respect to our geographic mix of revenues, U.S. revenue was $688 million, growing 19% and international revenue was $498 million growing 75% year-over-year. International revenue represented 42% of total revenue in Q2. Notably, this marks our highest international revenue contribution to date. Scaling our international presence remains a significant opportunity and a key strategic focus for the company, and we are very pleased with the team's execution this quarter.

Moving on to our balance sheet. Our liquidity remains robust with over $1 billion in cash and investments at the end of the quarter. Cash flow from operations was negative $136 million in the quarter primarily reflecting strategic component purchases made to support customer demand and secure component supply to fuel strong growth of our core business. These purchases consisted primarily of NAND and other key components that were intentionally made to mitigate the impact of continued cost inflation and further increases in component pricing. These actions are consistent with our long-standing approach to supply chain management and reflect prudent operational planning to meet the strong demand across our portfolio.

While these purchases created a temporary headwind to operating cash flow in the quarter, we expect operating cash flow to normalize over the next 2 quarters. Capital expenditures were $101 million, representing approximately 9% of revenue for the quarter. Our capital investments continue to support the continued scaling of our hyperscale business and to accelerate growth of our Evergreen//One subscription offering, while also reflecting the higher price of components. As a result, free cash flow was negative $238 million, and we expect free cash flow to track back to operating margins during the course of the year and expect free cash flow for fiscal year '27 to be between $600 million and $800 million.

In Q2, we repurchased 932,000 shares, returning approximately $69 million to shareholders. We also paid $74 million in withholding taxes on employee awards, offsetting dilution of approximately 1 million shares. We currently have about $176 million remaining under our existing $400 million repurchase authorization announced in Q4 '26. Finally, our head count increased sequentially by 282 employees, bringing our total head count to 6,900 employees.

Now turning on to guidance. I would like to take the opportunity to explain the rationale for our guidance raise. As mentioned in the prior Q1 earnings announcement on May 27, 2026, we were unsure about sustainability of demand as market participants had to adjust to unprecedented price increases that had not been experienced in years. Equally, the supply environment was tight and allocation-driven. The combination of these 2 factors led us to argue that it was too early to call for further upside to our guide in the second half of 2027.

Today, although market prices have increased to levels not experienced since 2017, demand remains strong, and we have anticipated continuous supply tightness with further strategic buys. Also, as we are past the half year point, we have now 2 quarters of visibility towards the end of fiscal year '27. With this new backdrop, we are now in a position to significantly increase our guidance for the second half of the year. For Q3, we anticipate revenue to be in the range of $1.325 billion to $1.335 billion, representing approximately a 38% increase year-over-year at the midpoint.

We expect operating profit to be in the range of $265 million to $275 million, representing approximately also a 38% year-over-year increase at the midpoint. The outstanding strength of our Q2 results, good short-term pipeline visibility and continued momentum we are seeing across our customer base gives us confidence in our full year guidance. Again, I would like to remind everyone for prior guidance that we continue to expect significant hyperscale product revenue in Q3 and Q4 based on order commitments through the hyperscale supply chain for our DFM solutions.

For fiscal year '27, we anticipate revenue to be in the range of $5.030 billion to $5.070 billion, representing at the midpoint, an increase of more than $500 million relative to prior fiscal year '27 guidance. In growth term, we expect revenue growth year-on-year to be at 38% at the midpoint. This is an increase of 75% in growth rate from prior guidance. We expect operating profit to be in the range of $940 million to $960 million, representing approximately at the midpoint, an increase of $110 million relative to prior fiscal year '27 guidance. In growth terms, we expect operating profit growth year-on-year to be at 50% at the midpoint. This is an increase of 54% in growth rate from prior guidance.

As Charlie mentioned, we look forward to providing you with an update on Everpure's long-term strategy, path to growth and long-term financial framework at our upcoming Financial Analyst Meeting that will be held on Wednesday, September 23 at our Santa Clara campus. With that, I'll now turn the call back to Paul for Q&A.

Thanks, Chuck. [Operator Instructions] Operator, let's get started.

[Operator Instructions] Our first question comes from Amit Daryanani from Evercore ISI.

congrats on the impressive print and guide over here. Charlie, I think the question you dealt with over the last 90 days is why are you forecasting, implying this big decel in your fiscal year guide. You clearly implying no more deceleration of this updated 37%, 38% growth on the guide. But -- and I heard all the reasons you kind of gave upfront and what's driving some of this trend. But if you could just help us appreciate when we think about the sizable uplift in guide against 90 days ago, you're taking it up by 16%, 15%, what specifically changed in the last 90 days? And is there a way to think about how big the factors, where maybe it's end market or better supply or pricing or new workloads, I think the guide is obviously very impressive in the step up. It would be helpful if you could just unpack what's driving it and the durability of of it versus 90 days ago.

You bet, Amit. And thanks for the question. Just to set the record straight. we were -- we repeated multiple times in the last call that we weren't saying there was going to be a deceleration. We simply said that we were not updating the annual guidance beyond what we had already seen. So it was the lack of an update. Now why is that? And you're absolutely correct, what changed?

Well, first of all, as you well know, and for people who've listened to this call all the time though, we generally operate with 1 to 2 quarters of visibility, not a full year. So we -- I don't -- I cannot remember a time when we raised guidance after Q1 other than just the performance that we had in Q1 plus Q2. So -- but things did change over the last quarter. So your question is absolutely appropriate. Two things went away. The concern about ability to source components to be able to deliver. And the second was the understanding of what our customers and market participants would do under the scenario of far hotter prices.

As you may remember, price increases were implemented in Q1 but really didn't fall into the customer environment until Q2. So this is the first full quarter that we've seen of higher prices. And we now have a much better understanding of how customers are responding to the higher prices. And we also have now 2 quarters of visibility. So 2 things went away, concerned about the higher prices and the effect, the concern around supply chain, which we feel now we are very much on top of. And then one thing, new information that came in, which is actual experience in how customers are responding to the higher prices. I hope that answers your question, Amit.

Our next question comes from Aaron Rakers from Wells Fargo.

This is Michael, setting off on behalf of Aaron. I also want to say congrats on the results in the second hyperscaler design win. For my question, I just was wondering if you could unpack sort of what's baked into the full year guide as far as the trajectory of product gross margins and OpEx growth as well to the back half?

You bet. So let me start with that. We -- I want to make it really clear. We are very intentional with our gross margins. I feel that we are in complete control of our gross margins, and we are choosing to operate at the lower end of our normal range in order to help customers -- to share the pain with the customers to help customers through this very difficult transition that is going on in the industry through no fault of their own. I mean the demand for semiconductors has outstripped supply that's driving semiconductor prices higher across the board from the smallest and the largest, most complex, and it affects us as someone who assembles those semiconductors into equipment.

Tarek, I'll leave it to you to the other part of the question on OpEx.

Yes. Thank you, Charlie. I would add to that, that I'd like to add some color on intentionality that Charlie has referred to a moment ago. Our goal is not to maximize gross margin in percentage terms. Our goal is to seize the opportunity to accelerate the growth and continue to gain market share. We are deliberately choosing to operate at the [indiscernible] of our product gross margin long-term range of 65% to 70%.

And this strategy is paying off. It's paying off visibly in terms of operating leverage, and you can see 2 things. Number one, the top line growth that we're driving, that I consider to be very impressive at 38%, but also at the line at the operating profit level, we're driving substantial growth in operating profit, 77% year-on-year because we could see that acceleration outpacing the acceleration of our OpEx cost in totality. So this is very intentional and it's about driving market share gains, revenue growth and operating leverage.

Our next question comes from Howard Ma from Guggenheim Securities.

I want to add my congratulations, too, on a spectacular performance all around, including the second hyperscale win and demand holding up in light of these unprecedented price increases. So my question is, can you help us better understand the ASP versus volume mix so far this year? In Q1, I think you guys said pricing and pull forward was roughly 10 to 15 points of growth, and therefore, volume growth was about 20 points of that growth, or I guess, a little over 20.

Volume was down in Q2, as you guys said, or I should say, more specifically overall capacity shipped. Can you share how much? And then with ASPs up over 100% already in the back half of this year, does that mean that your guidance despite the big raise that, that still implies a significant drop-off in capacity or a bigger drop off capacity back half versus Q2, and that's even after adjusting for the higher Evergreen//One sales? Or in other words, maybe a better way to ask it, too, is, is there still a similar level of conservatism built into the back half guide?

I'll leave that one for Tarek. So look, the way I like to phrase it is very simple, which is there is elasticity in the market. And when prices go up, elasticity works against the volume side of it. So the increase in revenues that we're seeing is not in proportion -- the proportion of total capacity shipped however you calculate that does not -- has not kept up, is far less than the amount of total revenue dollars that are shipped. And as we see prices start to stabilize, which I would hope to see towards the end of the year, I think we'll start to see volume start to come back up.

But Tarek, do you want to comment further.

Sure. Thank you, Charlie. And thank you, Howard, for the question. I won't comment on the second part of your question. I would simply say to you that there are 4 things that have driven our performance in Q2. Price, mix, unit systems and also capacity. So overall, system units are down. We see customers paying more for systems at the high end of our portfolio, that's also explainable by the mix shift and capacity in the systems that they buy has gone up.

So it is only the units of systems that has not grown. And that's why we have observed across our portfolio. And that is, in a sense, a very good results, considering that in Q1, as also was anticipated, there were some pull-ins and we were transparent in the fact that when you have substantial price increases, customers tend to bring forward some purchases and that's why we had pull-ins in Q1. But in Q2, this is no longer the case. Pull-ins did not play any role, and we don't anticipate any further impact from so-called pull-ins in the rest of the year. We feel very good about our guidance as it stands.

And I think that if you really look at our top line growth at 38% you would have to adjust that also for the growth of Evergreen//One. Evergreen//One has grown spectacularly well in Q2. As Charlie remarked, it's in excess of a $1 billion run rate. So our total growth once you normalize for the Evergreen//One contribution is well north of 40% approaching the 50s.

Our next question comes from Mike Cikos from Needham & Company.

This is Matt Calitri on for Mike Cikos over at Needham. We're curious about contract duration and deal size trends for Storage-as-a-Service customers that underlie the strong growth you're seeing there? Are you seeing any difference in how new and existing customers are approaching Evergreen deals, particularly as it relates to near versus long-term planning in the current supply environment?

Yes. No substantial change other than larger volume. Average contract length tends to exceed 3 years, somewhere between 3 and 4 years. So there's been no substantial change in that. What we are seeing is that -- well, first of all, just because of the way it works, we did not need and we did not raise prices in the in Evergreen//One as a service as much as we had to on product sales. And we've gone through why that's the case on multiple occasions. And what customers see is that it's a much -- they can -- they're able to land on lower prices over a longer period of time with the as-a-service environment, and they appreciate that stability.

So -- and we've seen this in the past that when prices go up, we see relatively more demand for Evergreen//One, which is great because part of the challenge in an as-a-service business when it's on-prem is that customers weren't used to it. So we really welcome the additional attention it's getting. But also when prices go down, then customers may arbitrage and look to buy a product rather than go with a service. So we'll see both sides of that.

Our next question comes from Erik Woodring from Morgan Stanley.

Congrats on the hyperscaler win and nice guide. Charlie, when I think about why Everpure wins some of the deals in these higher-growth hyperscaler environments, today, it then seems super cost-effective to replace nearline HDDs with DFMs, just given where QLC NAND pricing is versus HDDs. So in these new kind of wins and environments with hyperscalers, like what exactly are you displacing? Is that traditionally SSDs? Or maybe ask, what is the value proposition that these new customers are seeking with your DFMs that they couldn't get with their prior solutions?

Yes. So I'll start. I'm going to hand it then over to Rob, who works this extensively. But at a top level, you're correct. We are mainly at this point in time, replacing SSDs. Now our solution can be extended down to HDDs when the pricing comes back into line. But in the meantime, we replace SSDs. And you might ask, well, what is our benefit over SSDs? Well, quite simply, it's everything from power performance to absolute performance to ease of use because we have one software solution for any size, any scale, any performance level. Longer lifetimes, lower failure rate and then I'm going to hand it over to Rob.

Yes, Erik, just to add on to that, Charlie has enumerated a number of the benefits of the DirectFlash technology, which Obviously, our enterprise customers are well familiar with and have been able to be beneficiaries of for some time. Where the hyperscalers are now seeing a lot of promise for the technology is, as they're looking at, to your point, now some of these SSD estates, comparing the benefits they can get in those areas that Charlie enumerated versus SSDs.

And the dramatic effects that has on reducing their overall cost and just call it a take of operations, at hyperscale. When you think about -- just to put some numbers for comparison, when you think about the significant reliability benefits that DirectFlash drives versus SSDs, 5x more reliability on a component basis. When you think about the density benefits, the fact that we can be 2, 3, 5x is dense, when you multiply those out and you look at the overall reliability benefits that drives, you're getting into 25, 50x type of benefits.

When you look at efficiency when you look at power savings, similarly outsized amounts of benefits. And as compelling as those are in the enterprise, you now think about this at hyperscale and those benefits are very, very significant. So look, I think it's very early days. We're very pleased to be able to announce the second major design win. But more so, it's a great validation that DirectFlash software has a lot of room to run in this new market that we're opening up.

Our next question comes from Jason Ader from William Blair.

Just on the hyperscale business, do you plan on providing any disclosure on the size of the business maybe at the end of the year or sometime next year? And have you raised your hyperscale contribution in the guidance for this year versus what you had baked in last quarter?

So starting off with that on the -- on what's baked in. It's mostly baked in. There's been slight modification to it, but nothing worth calling out. It is on track. It should be stated it's very much on track according to what we had indicated at the beginning of the year. With 2 hyperscalers now, it creates a little bit of, if you will, greater confidentiality, but not a lot. So we're going to still stay fairly keep it fairly abstract in terms of guidance. We've indicated the rough size of exabytes that we'll be shipping this year. We said next year be multiples of that, and we're staying on that right now as the guide. But yes, we can't go further. Once the business gets to be bigger than it is today, hopefully, with more customers than we have today, we'd be able to then become a bit more specific in terms of the overall scale of the business.

Our next question comes from Simon Leopold from Raymond James.

Great. I wanted to see if maybe you could do a little bit of compare and contrast of the second hyperscale deal to the first. And what I'm sort of trying to tease out is, I recall that there was some degree of R&D work that needed to be done in prep work to ramp it. So what's different about this deal versus the original hyperscale deal? And how should we think about those compared to the others you're pursuing?

Yes. I want to identify one thing that was different in the -- if you will, the closing, if you will, of the deal, and then I'll hand it to Rob on the technology side of it. One of the things about this deal and the reason why it took longer than we anticipated was that this particular hyperscaler didn't want to award the design win until they also had a contract -- a supply contract in place. And so how we had strong indications of a design win quite some time ago. We didn't have an official one until we have the supply contract. So that was one of the things that was a about this one. So Rob.

Yes. And Simon, just to add on -- maybe to get back to your original question. Look, if I step back from it without getting into too specific details, what I'd say is that the general design the environment, the integration, the packaging of our technology in the second environment is substantially similar to the first customer. As with the first design win, the focus is initially going to be in the warmer and higher performance tiers, as we spoke about earlier during the call. But as well, this customer is very interested in over time, having this consistent architecture, be able to be applied across multiple and lower performance tiers as well.

And again, similar to the first customer event and what we're seeing in our ongoing conversations across the board, a very horizontal storage environments across all workloads, including AI, but as well, serving end customer platform as a service and infrastructure services. Always will be some fit-and-finish integrations as with any large customized environment, but what's really nice to see and validates our strategy here is that the core IP, the core packaging, the core implementation of solution very, very similar as well as not just the solution from a technology point of view, but a business pie point of view, all right?

As we have had conversations with the financial community over the last 4 to 5 quarters, as we ramped up our experience with our first customer, really nicely landed on a business model that worked well for both parties mutually taking advantage of our -- the hyperscaler supply chain capabilities and our contract manufacturer relationships.

Next question comes from Krish Sankar from TD Cowen.

Congrats again for the really impressive numbers. Charlie or Rob, kind of question for you, again, sorry, on the hyperscalers. Very impressed just on hyperscale. I'm just kind of curious, how much of this -- is there a way to figure out how much of this is actually going to be recurring revenues? And how confident are you given that multiples are ExaBytes, is there any correlation to NAND prices versus your hyperscaler wins? And any chance you can announce a third hyperscaler win this year?

Yes. So with respect to the -- there is -- NAND prices obviously have a very significant effect on the business, both in terms of where the technology is targeted and also, frankly, in terms of our -- because we do manage -- we're very deeply involved in the supply chains that deliver the NAND to the customer. So it also affects how we have to engage at that level overall. So I would say NAND prices do affect our business in a very significant way.

And Krish, since you will speak in the second part of the question, we're not going to be speaking about individual customers and speculating about timing of design wins moving forward. Now with the second customer with the repeatability, we'll be speaking about and looking at this product category as a whole moving forward.

Our next question comes from [ Joseph Cardoso ] from JPMorgan.

I'll echo my congrats here as well. Maybe can you just touch on the prebuys that you highlighted in the prepared remarks, like how should we be thinking about the duration or how many quarters of supply you've now locked in? And any thoughts in terms of how we should think about the implications for margins? Or is the expectation here that you can pass through pricing?

Well, Joe, thank you for the question, and welcome. I think this is your first call with us. So glad to hear your voice. Look, we executed those strategic purchases to continue to support the growth of our core. Our core is growing extremely solidly, and we have now sufficient supply coverage for the foreseeable future. I won't go into the deals because it is commercially sensitive, as you can expect.

We also now have reached a point in time where pricing has roughly caught up with the cost of raw materials, and this is really helping the management of the gross margins moving forward. But like I said to your colleague earlier on the call, we continue to want to operate at the low end of our product gross margin range of 65% to 70%. That is to continue to accelerate top line growth and grow market share and drive higher levels of operating leverage. Hopefully, this helps.

Our next question comes from Wamsi Mohan from Bank of America.

I think you just noted in the Q&A that you were waiting for a hyperscaler customer to secure supply. And I'm wondering sort of a follow-up to your prior question, would you be willing to share if you are finding any long-term agreement with memory makers? Or is there an appetite to do that at all given that your hyperscaler customers are trying to secure supply? But for your non-hyperscale business, obviously, you've bulked up some supply with inventory buildup. But just curious if you're motivated or think that it might be necessary for you to sign any LTAs as you look out further beyond the next couple of quarters?

Wamsi, it's Tarek. Thank you for the question. We executed those strategic buys for our core, not for the hyperscaler business. We do not procure the NAND for the hyperscaler business. The hyperscalers procure the NAND for their business through their supply chain. And what Charlie was referring to before is, as we executed the second agreement with the hyperscaler, the hyperscaler was waiting to secure its own purchase agreements before signing the agreement with us. That is the distinction. We don't procure the NAND for hyperscalers. So strategic buys are just there to continue to fuel the growth of our core, which has been outstanding in Q2, and we have really good confidence for the rest of the year because of that.

Our next question comes from Asiya Merchant from Citi.

Great. Charlie, if I may, I think the commentary suggests that you have a good visibility beyond just fiscal '27 that you guys are guiding to here, but it seems like the commentary suggested that you have ability of continued share gains and durability of demand. So if I could just ask you to just unpack. Is it enterprise AI adoption that's really driving this and you see the demand towards these high-performance flash that's driving sustainability, just even beyond the hyperscaler business, if you could?

And just if I may, on subscription margins, maybe if you could peel the level here as well on when to expect subscription margins to come back to ranges that they were in the past?

Yes, thank you for further questions. So we -- as I mentioned at the beginning, we generally operate on about 2 quarters of good visibility of good pipeline that we -- that allows us to forecast roughly 2 quarters in the future. And so that's what we're doing right now. Now you're correct. I'm referencing that I believe that our momentum has increased and that we're picking up share faster than we have in the past or at least in the most recent past.

And because of that, we're more confident as we go into FY '28 because the momentum that we have, we feel is much stronger than we've had in the -- for multiple years. And the reasons I gave at the very beginning of my opening remarks is the basis of that. That is to say that we've completed building our products across the entire application use case space of data storage. We're leading the industry. The Gartner Magic Quadrant is another example of that. We continue to expand what customers can do with their data and how they can protect it, and how they can leverage it for AI. And we think that is benefiting us with outsized market share demands.

And that momentum, we think, will carry going forward. But I wouldn't be very clear that on a forecast or guide basis, we really only have about 2 quarters of visibility generally, and that's what we guide on is what -- that is our best prediction of the near future that we can have.

Sorry, let me answer the second part of the question on subscription gross margins. Look, there are many small factors that have -- that are behind the performance in this quarter. But the main one is being price increases in components, replacement costs for our forever program. And that is a good thing. And the reason why it's a good thing is that what we know from customers who take on forever is that their churn is lower. And so this helps in the long run our business, and we will continue to make investments to fuel our Evergreen forever program.

We're a little bit over time, but we're going to squeeze in one more question.

Our last question today comes from Tim Long from Barclays.

I think we covered a lot of this stuff here. I did want to -- maybe if you can just touch a little bit, Charlie or Tarek, on the international performance, really a big spike in the quarter. So just curious, I know that's been an area of focus. I mean, last quarter it was down a little bit. So I'm sure there's some lumpiness. But what are the kinds of dynamics there? Is this go-to-market? Is it better product availability? What kind of drove that and how sustainability -- how sustainable do you see that theater?

Well, we believe that we're going to -- our -- we've been operating at roughly a 30% plus international share of our total revenues, which is a bit low for -- and it's an area that we've been investing in for a long time. So we do expect that to continue to go up. But I will say on a quarter-by-quarter basis, better measured on an annual basis, quarter-by-quarter, remember, it's all based on shipments rather than on bookings and both shipments and bookings vary over time. So it really is better measured on an annual basis. But yes, we hope to see that continue to increase over time.

Before we conclude, Charlie, I think he has a final comment.

Yes. Before we close, I do want to thank our customers, employees, our partners, our investors and our suppliers through -- going with us through this very challenging period of time. Our continued success is made possible by your trust and your commitment. So thank you all very much. Speak to you next quarter.

That concludes the Everpure's Second Quarter Fiscal 2027 Financial Results Conference Call. Thank you for your participation. You may now disconnect your lines.

All right. Perfect. So afternoon guys. Day 1 here at the Morgan Stanley TMT Conference. My name is Erik Woodring. I am delighted to host Everpure CFO, Tarek Robbiati today for a fireside chat.

Before I start, a few disclosures here. So please see the Morgan Stanley research disclosure website at www.morganstanley.com/researchdisclosures. If you have any questions, please reach out to your sales representative.

And then from the Pure Storage side, statements made in these discussions, which are not statements of historical fact are forward-looking statements based upon current projections. Actual results could differ materially from those projected due to a number of factors, including those referenced in Pure Storage's most recent SEC filings on Form 10-Q, 10-K and 8-K.

So for anyone that doesn't know Tarek, Tarek joined Everpure last year. He brings over 25 years of financial, strategic and transformational leadership experience. Obviously, coming from HPE, but also RingCentral, Sprint, et cetera. So Tarek, thank you for joining us today.

Thank you, Erik, for having me. I'm delighted to be here with you and so with the audience.

So maybe before we get into earnings and everything that comes after that, you are no longer Pure Storage, you are Everpure. Help us understand kind of the significance of the name change. What that means and kind of what you're trying to message to all of us with that change?

Well, thank you for the question. I think the name change fundamentally represents the evolution of our strategy. And if you really want to understand what the core idea is we firmly believe that the future is about the data and data becoming the most important asset that any company has. And therefore, for us, as a provider of data solutions, it's not about just storing bytes of data, it's about making sense of the data. And we are moving into this direction because of the world we're living in, particularly with AI. And there is a greater and ever greater need to make more sense of the data and prepare the data for ingestion by AI models. And this is why we also concurrently acquired a company called 1touch, which is all about data contextualization, cataloging and making sense of it.

And so that's were we're moving. It's an evolution of what we do. And also, it is ingrained in the strength of our past. We didn't drop the word pure because we are a pure flash company. We also leverage the word ever because of our evergreen model and the fact that our solutions are constantly upgraded for customers and we felt it was right to combine the 2, and we now call ourselves Everpure.

Perfect. So let's follow up with that. And kind of talk about earnings that you just reported last Wednesday. You ended the year on a strong note. You guided to -- and I'm using the midpoint here, 18.5% year-over-year revenue growth, 26% year-over-year operating income growth for fiscal '27. Both of those an acceleration from fiscal year '26. Can you just unpack for us at a high level and we'll then get into the details, but at a high level, what exactly is driving the acceleration in revenue and profitability growth this coming year?

So thank you for the question, and thank you very much, Erik, for noting that it is an acceleration relative to our past fiscal year, fiscal year '26 that ended on the 31st of January. For those who don't know, our fiscal year starts February 1 and finishes on the 31 of January of the following year. So we finished fiscal year '26 extremely strongly. We had a wonderful fourth quarter. Our top line was at 20%, our operating margin was at 21.3%. By the way, we touched Rule of 40 in a quarter, a data point doesn't make a trend, but it's pretty good in my book because when I joined the company, I could see a path to that and we are on our way. But that has to be grounded in the strength of our core business.

And our core business is really doing extremely well, and there are many elements that contribute to it. First and foremost, we are positioned in the portion of the storage market that grows the fastest. So if you really look at the storage systems market it's about $60 billion, 50% of that is all flash. And for the first time, at the end of calendar year '25, all-flash became the dominant segment of the systems storage market. And that is growing at about 8%. Hybrid flash, which is a combination of hard disk and flash is growing at 0.5%, and then hard disk drives are on the decline. And this is for the next 4 years, and the numbers I just quoted are not our numbers, they are IDC numbers, and I'm sure you can verify those.

So we are positioned in the highest portion of the market that is growing the fastest. And we are the company that has always looked at storage as a high-tech solution to provide to customers, not so much a hardware commodity. And that has to do because we go and tackle the market by way of our software. So our -- now you know that, that's a market that you know a little bit about our business and how it's performing. In our earnings announcements, we discussed the strength of our core. We're moving from what we were traditionally known for as some -- a company that focused on commercial segments to more and more into the enterprise. And I gave snippets of the quality of our execution there. Our enterprise deals in excess of $5 million grew 80% year-over-year.

So we're really making headway, really moving forward into the enterprise segment, and we're very, very pleased with that. And it's actually -- you would imagine it's actually natural. If you want to scale up the business, you have to go for bigger deals. It's -- the signing of a small deal resource-wise and time-wise, roughly is equivalent to the signing of a bigger deals. You might as well go for a much larger ticket. And that's why we're doing. We continue to grow across commercial, we grow across enterprise. And also our offer is becoming much more sophisticated. We are bidding for larger portions of the estate in enterprises.

So don't expect the number of our customers to grow. We have 14,500 customers worldwide. It's not the way we intend to grow. We intend to grow in the right category, in the right segment of customers, much more higher tickets in the enterprise, and that's doing very, very well. So key growth in there, Erik. And then, of course, I'm sure you will ask about it. There will be growth coming from hyperscalers. We've been generating revenues with hyperscalers in fiscal year '26 for the first time. We did very well as far as I'm concerned in that segment. It's a very different sale.

It's not a system sale. I want to remind everyone. And for those who don't know, that we don't sell systems to hyperscalers for their production environment, we sell the software that enables them to manage the NAND that they procure through their supply chain. We'll come back to that for sure. But we do expect, and it's baked in our guidance, a very strong growth from hyperscalers this year, accelerating from fiscal year '26. And also, we are very confident about the prospects of that business for fiscal year and beyond. So hopefully, that gives you a high-level view of the sources of growth, Erik.

Perfect. Yes, we're taking notes here while I'm asking questions. I want to get the memory questions out of the way first and foremost, just get them out of the way, so we can talk more company specific. So first, kind of a once-in-a-generational memory cycle just at a high level, as you guys are thinking about that, what's the impact on the storage market? What's the impact on your customers and pricing elasticity, again, from the perspective of pure -- excuse me, Everpure.

Everpure, thank you. So obviously, we all know that with the unprecedented build out of AI systems, there is an imbalance between the demand and the supply of certain components. And this affects the storage industry and the compute industry to different degrees. It really hinges in my mind on the value add that each provider who plays in the industry provides to their customers. And if you think about our position at Pure Storage, we are a software company. We -- yes, we do sell appliances and we manage those appliances by way of software, if you want to put it very, very simply. But it affects us less, the movement of commodities affects us less than other players in the industry. And this is really, really important to understand.

So why does it affect us less? It's because our margin structure gives us a lot of leeway, if you want. We have 71% gross margins driven by our software. And you know, because we reported that we practiced a 20% price increase across our portfolio, on average, okay? Some prices went up more than 20% others went down, went up less than 20%. And so if you do the math, if you really look at a system, let's say, costs $10,000, now it costs $12,000, we protected gross margin. The gross margin remains the same as we price because we price rationally to protect gross margins. Then you have to ask yourself, knowing the increase in input cost in commodity prices, what is the proportion of our cost of sales that is affected by commodity prices going up.

And if you do the simple math there, you will see very little because of that 29% that represent cost of sales, a portion of it is support or subscription, if you wish, subscription cost. This is labor. It has nothing to do with commodities, and the rest is obviously commodities. And in terms of magnitude, commodities can increase quite dramatically. We can absorb those increases or decide to price them into what we offer. So you will -- if you do the math again, conclude that -- the proportion of our cost of sales that is subject to commodity price increases is relatively small.

And this is the reason why we're more immune to commodity input costs rising than the rest of the market. It has to do with our software. And therefore, it has to do with our margin structure. The second observation I would make is even though the price hikes or the input cost increase has been sudden and of a high amplitude, it doesn't immediately affect our P&L. The reason why it doesn't is that we have acquired commodities over the past few quarters at reasonable costs. And those commodities are blended with higher cost equivalent commodities. So this blend effect materializes in our P&L, just like on the revenue side, you have new pricing blending with old prices.

And so over time, this plays out to find a new equilibrium. So for us, commodity increases matter relatively given our margin structure, and we can navigate this period as long as commodity costs are stable. It's input cost stability matters more than the hike itself. And even if there are further increases in commodity costs, which we cannot exclude for obvious reasons, we can always change our pricing again, and I won't exclude that. And operationally, we can change our pricing over a weekend across the entire portfolio. So we can very quickly react to the change, and we monitor this very, very rigorously.

The other observations that I want to highlight to you is why were we able to only increase prices by 20%, where you know that the competition has increased prices 35%, 40%. It has to do with what I discussed with you before, which is the proportion of our cost of sales that is impacted by input costs. And we were able, therefore, to also increase prices. We were the last player increasing prices because we wanted to see what was going on. So that gave us an advantage from a timing standpoint. I would also say to you that other conditions attached to orders are really important. So how long are quotes ready and available for Erik. Our quotes are available for 60 days, unchanged.

And we honor the quotes for 60 days. This matters a lot for customers. Because if you ask what other players are doing in so far, validity of their quotes, some of them have quotes that are valid for 14 days. Now that is -- I would not advocate to have quotes valid for 14 days for a very simple reason. If you're in sales, it would be your worst nightmare. You have to re-quote every 14 days. And if you re-quote every 14 days, the first quote you're going to try and do is going to be the one that is going to be accepted. Therefore, you're going to be not necessarily optimizing the margin as you sell. So that's a very important consideration to be had.

The other operational consideration to make is also in terms of our delivery time frames, we have not seen a substantial elongation of delivery time frames. We deliver somewhere around 45 and 60 days. It may be a little bit more, but it's not something that customers are not willing to accept. And I want to finish with -- on this question about customers in saying, look, we have an 84 NPS score. This is not common in the industry. And we are very jealous about it. So we work with our customers. We don't gouge our customers in circumstances like this. And we have a business model with Evergreen. I'm sure you would want to talk about that, that allows customers to make different decisions as to how they want to acquire solutions for the data.

And extremely thorough. So thank you for that, Tarek. Just one point to follow up on quickly before we move on is just the point of supply, just the concerns that some would have on second half supply availability. Recognize you talked about acquiring commodities in advance over the last few quarters, just comfort around the ability to procure supply.

Look, we do have long-term agreements with suppliers with whom we partnered over time. Suppliers are as important to us as customers are. We work with them, and we have consistently worked with them over time. Those agreements helps us mitigate but do not eliminate supply risk. I won't say we are completely immune, but we are probably less susceptible. But nonetheless, nothing comes for easy in this environment. We have to work hard to make sure that we can stand by our delivery deadlines and provide the services and the products to the customers as they want.

Okay. Perfect. So when I kind of go back and dissect your '27 outlook, kind of back out the hyperscaler, at least our estimate for the hyperscaler. You are embedding a pretty significant deceleration in what I would call the core of your business. And so I guess the question is how much of a result of that is kind of pull forward and maybe not so much certainty in the second half of the year versus hey, there is uncertainty. So we're just being conscious in taking kind of a prudent approach and making sure we don't kind of overstep because there are still questions about the second half.

Yes. I really like the way you've asked the question, Erik, because it's really very, very much balanced. A couple of observations. First, we finished Q4 very, very strongly. Our top line grew 20% or in excess of 20% and the linearity in the quarter was back-ended. I mean I was not expecting to finish January 2026 as strong as it is. That's the honest answer. And so you know that we book orders and then we recognize revenue upon shipment. So most of what we booked in January and a few weeks before that, in December, will shift into fiscal year '27, Q1. And so that's why we guided -- as we guided for Q1, we are at the 28% growth in Q1 versus 20% in Q4. So if you would like to estimate the amount of dollars that were pulled into fiscal year '27 from orders placed at the end of fiscal year '26, a good estimate, a reasonable estimate would be to look at the difference in growth rates.

So 20% for Q4 and 28% for Q1. That's 8 points more or less majority of that is pull ins. So even if you were to take the conservative view and say 8 points is all pull-ins, that's $80 million. It's not an extraordinary heavy amount. The demand remains steady beyond that point, and there is a rebalancing of the demand as the new pricing takes hold. So we also have to be -- answer your second part of the question, Erik, relatively easy compare in Q1 over Q1. And we also have tougher compares Q3, Q4 '26 on Q3, Q4 '27. And so that is also part of the equation.

And finally, the third one, allow me to say it because I have to say it, there's an element of the tyranny of the spreadsheet here. And what I mean by that is what we've seen happening is the fact that, of course, the street locks in the Q1 estimate, locks in the Q4 estimate, a full year estimate and everything that is in between is derived by difference. Correct. Arithmetically correct. I can't hold anyone for that. I would simply say don't make the wrong inference about it. Right now, as you pointed out very well, I don't have visibility on what happens in Q4. It doesn't mean that it needs to be interpreted as I don't know what's going to happen in Q4. I am casting out on what's going to happen in Q4.

It simply means I don't manage the business that way. We have visibility a couple of quarters-and-a-half, maybe 3 ahead, but we know how to solve where we get to, and we have a 2,000 strong sales force that is there to really work and find new opportunities and get those opportunities materializing in the second half of the year. I am only in my first innings of fiscal year '27 and still have a long way to go. So far so good, we stand by our guide for Q1 and obviously, our guidance for the full year.

Okay. Perfect. Let's transition to FlashBlade//EXA, new high-performance, parallel processing, you're targeting Neo clouds, GPU clouds. At earnings, you highlighted your first major win with EXA. Just like how long did it take to get that customer? Is there a way that you can help us understand the size of materiality or even sustainability of that win? And then you mentioned in your earnings call you got the customer to switch from an alternative vendor. What did you do to get that switch? That's like a 4-part question, but...

Yes, yes. No, no. But it's excellent, it's perfect. It's very comprehensive. So thank you one more time. So EXA is a brand-new offering. It's a high-end architecture aimed at typically Neo clouds and larger scale deployments in the enterprise. It was made -- I was reminded this morning that it was made available in general availability in June 2025. And so from general availability to the sale, we had 6, 7 months that passed, and we were able to have the first win with EXA. And yes, you're correct. We were able to persuade a customer to use us and switch from an alternative provider because what is very apparent are 2 things. One, the performance of EXA is materially better than anything else available in the market. Is visible through third-party benchmarks like the MLPerf publication, which shows that we are 2x faster than the next competitor.

And two, it's also the ease of deployment. That plays a huge role when you consider those Neo cloud, we have to put online, an enormous amount of infrastructure and fire it up and make it working. So that was a key differentiator. Now we have a few things to do for EXA to really take off. We have a couple of features that we're missing that we have to build. They are on the road map. And we are confident that in fiscal year '27 will have further acceleration of EXA, everything that we have now in mind expectation-wise, is factored into our guidance.

Okay. And just a quick follow-up. On the call, you also mentioned dozens more in advanced solutions again, just framing if we take a big step back or even multiple years of step forward, the opportunity size for kind of this TAM, which is clearly kind of incremental, what I would say, to your core 60 billion systems TAM that we talked about.

Yes. Yes. So I would say directionally, it's -- we believe it's a large opportunity, probably not as large as hyperscalers, but large nonetheless, and it's good so.

Perfect. All right. So let's move to the hyperscaler side of the story. Obviously, you've guided to your first major hyperscale customer ramping pretty significantly in the second half of this year. I would love to better and maybe we would love to better understand just like the sustainability of the growth story with this customer, not -- feel free to give us as many numbers as you want to, of course, but just how do we think about the sustainability beyond what you've kind of already outlined for fiscal '27?

All right. Let me try. So in fiscal year '26, we -- it was first year where we're generating revenue. And I think your own estimates probably speak about a few tens of millions of dollars there, which is, I would argue, pretty good for a business in the first year of operations. We do believe that this year, fiscal year '27, we will accelerate that quite substantially, and we are very confident about the outlook in fiscal year '27 and beyond fiscal year '28. And there are many reasons for that. One is obviously need. And we discussed that over our earnings call, Charlie, our CEO said, it's obviously a bit of a tailwind that we have a bit of a crunch in commodities, which is making it helpful for our conversations with hyperscalers.

But every vendor is also having to reckon the dependency on commodities and therefore, in the short term, we'll have to navigate that. Having said that, if you really think about what a hyperscaler buys for their production environment, they buy 3 things: hard disk drives, SSDs and now they buy Everpure DFMs. Capacity in hard disk drives is not going to increase. There is no rational business case one can make to say we're going to put more capacity on hard disk drive. The reason being it's a technology that is way too old, way too inefficient from a performance standpoint and increasingly costly from a cost of ownership standpoint. Then you're left with SSDs who are obviously better than hard disk drive. They use NAND.

But they are also a mini system by themselves because there's a CPU, there's control as a whole heap of things that involve their own technological overheads, and therefore, an SSD doesn't exploit the full capacity that the NAND it is built on provides. And so if you are a hyperscaler today and you have those extraordinary buildouts, you're keen on finding a third solution. That's where we come in. And we feel very good about this because we're gaining more and more acceptance as an alternative in our conversations with hyperscalers and it's just a matter of time to break through that. And we've geared ourselves to do so.

So let's talk about that. The -- I hear a ton of confidence from you guys about the hyperscaler pipeline. The confidence in expanding that customer base, maybe the question is, what are you finding is the biggest friction point in getting those types of customers over the finish line? Yes, maybe let's just go with that. Where does the confidence come from? What's the biggest friction point?

Brilliant question. So to make it work, we have to solve 3 dimensions. The first one is the technology and this is what we call the qual process, making sure that the NAND that we provide in the hyperscale environment is qualified and tested to work with their software, the binary software that they use in their environments. And that is an incredibly lengthy technological test process. It's akin to testing a plane before its first flight, right? That's effectively the analogy I could best give you. So the amount of testing that goes through is very, very thorough and lengthy. And once accepted, then you make it into the next generation, the next road map of the hyperscaler for data centers, their data centers. That's the first thing we have to crack. But in many of the conversations, we made great progress.

The second thing is the business model. And to remove the friction, as you say, we've simplified and standardized the business model. We provide our software to manage the DFM. We also provide some componentry that effectively is -- that provision of that componentry is designed to eliminate friction points, drive greater adoption by hyperscalers of our solution overall, right? So the only thing that hyperscalers have to worry about at that point is the procurement through their supply chain of the NAND, right? And then effectively, everything else works fine. And the economic implication of us providing that componentry is that we will generate relative to what we started doing the same gross profit dollars, but that componentry comes at a very small gross margin. We have probably a little bit more revenue and practically unvariable -- invariable gross profit. That's why the gross margin ranges from 75% to 85% versus the 90% we flagged in the earlier model. So the model standards, we provided componentry. It's just really to drive the ease of use and adoption by hyperscalers.

And the third thing we have to crack is everything else that is operational, for example, support, who provides support, what is our role in providing support, spares. Do they require spares or they not require spares and so on and so forth. And that's the third bit we worked through in the past few months, and we found solutions for it. So now really, we have made it a lot more standard to move forward with us. The continued long haul and intent is testing, but we're confident we will crack that moving forward.

All right. Perfect. I want to talk to gross margins quickly, which is and specifically, product gross margins. You've kind of talked about the improvement in -- we should expect an improvement in gross -- product gross margins through the year. Obviously, the ramp of the hyperscaler business in the second half helps that is accretive. If you took that away, how do we think about the product gross margin for, again, core product? Or what gives you the confidence that you'll be able to improve those product gross margins through the fiscal year, just in light of the headwind and some of the things we've talked about on memory?

Yes. So our product gross margins are estimated to range from 65% to 70%, excluding hyperscaler, excluding also Portworx term license revenue. And so what gives us confidence that we can see them improve. And by the way, at the end of the fourth quarter, they were at 67%, so we are pretty much in the middle of what we said we would be. We do believe that, as I explained before, pricing will take hold and that also we will have a better matching of commodity costs with new prices. But also, we have other levers to pull. So remember, we provide a number of products, and they all have different degrees of commodities built into them.

So product mix is a very important lever we can pull. High-capacity products in the solutions we provide have more componentry than others. High performance have less and so on and so forth. So we can pull the lever of product mix to really make sure that we tilt it towards the right volume mix in favor of higher gross margins. That's also a very clear lever we can pull. And we've done so in the past and we continue to do so moving forward. So these are the things that give us confidence beyond the fact that we do have Portworx revenue, and we also have hyperscale revenue.

All right. Perfect. Last 2 for me before we wrap. Just free cash flow for a long time, free cash conversion was greater than 100%. It's dipped closer to 90% the last few years. Does that return back to 100% plus? Has something structurally changed where there's maybe working capital investments you need to be making just understanding kind of where that should go from here as you ramp new businesses, touch on new customers, et cetera?

Yes. So our free cash flow tracks very well. Our operating profit margin is between the free cash flow margin and the operating profit margins, there are 50 basis points difference in favor of the operating profit margin. That's what you can expect. And it's really pleasing to see the free cash flow growing the way it has. We are also within that calculus. We're estimating CapEx to be between 7% and 9% year-over-year. There are things that we're investing in -- in R&D and in various aspects of the company as we continue to grow and expand. But expecting free cash flow to be in line with the operating margin with a small delta is probably a good assumption moving forward.

Okay. Anything on capital allocation that's changing in terms of priorities as we just look ahead into fiscal '27?

No. Pretty much the same. We have a buyback program. We have about $300 million on that buyback outstanding. We -- if you just take the effect of the buyback on dilution, we've offset about 56% of the dilution, thanks to the buyback program. We're roughly at about 50 -- some part, I mean, we dedicated 56% of our free cash flow to the buyback program, which is roughly what we flagged. On top of that, we do withhold to cover for stock-based compensation, which offset the dilution even further. So no change to the capital allocation that we have practiced in the past, and we'll continue like this moving forward.

Okay. Maybe last question as we wrap, just a final word. You could give us what do we not appreciate? What do people not understand? What are you excited about? Kind of you get the last word here.

Well, thank you for giving me the opportunity just as the clock was passing 0, which is great. So I'd say to you, look, I know there is a tremendous focus on hyperscalers, and I completely understand why. But please do not oversimplify our growth and neglect the growth that happens from our core. Our core is humming. If you really put things in perspective over the past couple of years and what we guided this year, we started last year guiding 11%, finished 15%. We're guiding 18.8% for this fiscal year. The vast majority of that growth comes from the core. And we are the share taker in the industry, and we will continue to be so because we have a -- we believe with our software, a significant advantage.

Perfect place to end. Thank you, Tarek.

Thank you, Erik. Thank you very much. Thank you.

Thank you.

Okay. So hopefully, I counted 321 correctly. So warm welcome, everyone, for our fireside chat with the team from Pure Storage. Before we get started, I would like to read the disclosure. Statements made in these discussions, which are not a statement of historical fact are forward-looking statements based upon current expectations. Actual results could differ materially from those projected due to a number of factors, including those referenced in Pure's most recent SEC filings on Forms 10-Q, 10-K and 8-K.

So to that end, I want to thank everyone for joining us, a perfect timing given for Pure's recent earnings call.

I do have some questions, kind of follow-up, and I'm just going to start. Rob, you guys talked about hyperscaler revenue that is embedded in your fiscal year guide, especially for the second half. What I want to better understand is the diversity of customer. Is that just a one hyperscaler customer, too? And then the mix, does it include both hardware and software?

Yes. Thanks for the question, Mehdi, and thanks for having us, as you mentioned, coming off of a great 4Q print and had a lot of discussions with the financial community over the last couple of days. So let me hit your first question, which is, I think, just seeking to better understand the hyperscaler revenue expectations, where is that coming from? How does that look? And then a couple of elements in what we introduced as the new more standardized hyperscaler business model.

As you mentioned, we discussed on the call on Wednesday, our increased expectations for the year relative to our prior commentary in terms of hyperscaler shipments, that's -- as we have tried to share in qualitative color, that relationship continues to go very well and is ramping quickly and has surpassed our prior expectations. We do expect most of those shipments in FY '27 to occur in the back half, so in the fiscal 3Q and 4Q just due to the schedule of the hyperscaler build-outs.

To your question about customer mix and concentration, yes, at this point, we are still shipping to our lead customer in the space. We are engaged with -- continue to be engaged with in advancing engineering discussions and I would say, exploration and testing with additional future prospects, but really just one primary customer in the scaled part of the hyperscaler business at this point.

And then I think the other part of your question, and I know I'm going along here, but you had 3 questions baked in. I think the other part of your question is seeking to better understand the mix of that revenue in terms of, hey, how much is software, how much is hardware? What does that exactly look like? So I want to take you back to and take your listeners back to our original business model, which we articulated in fiscal '26. And in our 3Q call, we had kind of foreshadowed and indicated, hey, we're going to be evolving this business model and then take you to what that looks like and our expectation moving forward and what would cover most of the FY '27 shipments.

So the original business model, if you recall, was really centered around -- think of it as a software royalty or license fee for the direct flash software. That is really the heart of the IP. That's the heart of what the hyperscalers are interested in. And in FY '26, that was -- that made up the bulk of the revenue. A smaller piece in terms of maintenance and support, but we can set that aside. Really, the bulk of that was in software royalties. The hyperscaler supply chains would then in order to take advantage of that software, hyperscaler supply chain would be procuring the servers, the DRAM, the NAND itself, all the components to make the NAND into the direct flash modules, et cetera.

What we learned as we've grown the relationship and have talked to additional hyperscalers is hyperscaler supply chains are quite adept at procuring servers in DRAM, quite adept at procuring NAND and NAND flash. For all of the other components to finish out the direct flash module, our PCBs, our designs on flash controllers, et cetera, it just made sense for all involved. It was a lot operationally simpler for our supply chain to go help procure those components. To be very clear, and we discussed this in Tarek's prepared remarks, we would not be procuring the NAND for the hyperscalers or the servers, really the rest of the components to make that solution.

And so when you consider that in the overall shipment, that is what -- that change or that shift to that standardized business model is what is now bringing the gross margin expectations associated with the hyperscaler business into that range of 75 to 85 points. And depending on the configuration or the performance tier or the drive sizes, you may have some variation within that range, but that would be our expectation for that hyperscaler business.

That's pretty, I would say, clever, let your customers and their scale be involved with NAND procurement, which hasn't really happened in the past. I guess, perhaps in the past, most of the demand was driven by enterprise. This is -- the AI has enabled you to penetrate hyperscalers, and now they are helping you with procuring the required NAND. Would you agree with that?

Well, so yes, let me maybe unpack that. So obviously, there's a lot of complexity that goes into sharing and coordinating supply chains. And as you mentioned, we -- for our enterprise customers for our core business -- our customers are just looking for the best solution, finished product, et cetera. So it really doesn't make sense and they don't have the infrastructure in place to go run these types of quite sophisticated supply chains. Hyperscalers, on the other hand, it makes sense to -- for us to work together with them to apply the best capabilities they have in their supply chain, the best capabilities we have. And that's really what led to this evolution and standardization of the business model.

Yes. Now I want to -- just about of the queue I have with my questions and topic. And actually, go to the demand side, and there are evolving different drivers. How do you see KV caching impacting the demand, especially as we migrate to inferencing? And what are the key products in Pure's portfolio that will be applicable to that, which I would phrase as a new kind of a demand driver, the KV caching?

Yes. So if I step back and look at the -- now shifting gears to the core business, if I look at the demand drivers on the core business, I think we see a number of demand drivers we talked about on the call and reflected in our guidance and outlook for the year. We see robust demand across the board and exiting fiscal '26 with a ton of momentum. And I think there's a number of demand drivers across the industry.

And then I'll answer your question specific to AI, which, of course, is still a smaller part of our overall book of business. I think that certainly if I look across the industry, what we saw in -- throughout the back half of fiscal '26 you saw us raise and subsequently beat the guidance several times through our fiscal '26. What we saw was a couple of things. One was a pretty robust demand across the board across segments, whether that was enterprise, commercial, U.S. international, it was pretty broad-based. And I can't identify specific application driver or specific vertical or area. We saw pretty broad-based demand, and you saw that come through in terms of our sales and our revenues as well.

I think I suspect part of this may also be last year, right, we had a little bit of, call it, a headwinds to IT budgets across the board, perhaps exacerbated by the Broadcom VMware action. A lot of folks were pretty stretched, and you see some of that either pressure relieving, if you will, or maybe folks that had sweat assets a little bit longer. You see some of that starting to kind of relieve. I think the other thing is, as folks now turning more in the direction of AI, I think as folks are progressing in their thinking and plans about, hey, how do I go get more value out of my data modernization -- infrastructure modernization projects overall. We're seeing those move forward. And we see that both in the core array business, array and blade business as well as what we're seeing in Portworx in terms of modernizing not just container applications but virtualization as well.

If I look specifically at AI coming to your question about KV cache, we're engaged in conversations across the board, everywhere from folks that are just looking to connect their data to chatGPT all the way to folks that are doing some small-scale maybe fine-tuning, maybe folks that are doing -- really focused on inference and RAG and to large-scale GPU clouds or foundation AI natives that are really down the training path. I think we have a robust portfolio of solutions for that.

On the inference side, we do see more of a focus on KV cache. I think it's very early days. We have a number of customers on our FlashBlade solution that are making use of our object capabilities and our KV cache accelerator there to really make that whole process a lot faster. I do think as inference grows to scale, that will become more and more important. But as a portion of the market and just overall segment of kind of a share of our book of business, it's still relatively small compared to everything else that we do.

Yes. Yes, especially your NAND suppliers are having a hard time sizing the market -- the size of the market for KV cache and so everybody is trying to figure this out. But it's very fascinating because you have the enterprise finally beginning to utilize AI and then you have the native data, the hyperscalers wanting to own everything. They have already invested billions -- hundreds of billions in training and now they want to extend their services to inferencing. So in that context, how do you see neocloud service providers that are telling enterprises, hey, I could be your edge compute evolving. When you say enterprise demand is strengthening, is that traditional enterprise? Or does that also include neocloud service providers?

Yes. So a couple of pieces to the question there. When we talk about demand strengthening and just being robust across the board, really talking about traditional enterprises, traditional commercial, public sector, really our core business. And again, just highlight that, that we do see that pretty robust across all workloads, of which AI is just one of many.

In terms of how our -- where do neoclouds fit into the equation? It's interesting, they -- I think we work with many neoclouds. We work with them on behalf of both, call it, large-scale tech titan types who are utilizing their capabilities as well as smaller enterprises and commercial customers that are utilizing them to provide the GPU infrastructure. I think the neoclouds are really focused on -- they're focused on providing infrastructure that can meet the wide range of demands. Think about it this way. If you're a GPU as a service provider, a neocloud, you might have one customer who's doing super large-scale foundation model training. You might have a couple of other customers that are just doing inferencing. You might have a couple of other customers that are actually working with images and video instead of text.

So the range and variety of different needs that you have to go satisfy is quite large. And I think that presents an opportunity for us because unlike most of the storage competition in the AI space, we have not taken -- we've purposely not taken a laser-focused view, right? We've leveraged, we've really leaned into the power of our core technology, which is to be very flexible, very adaptable, very deployable for multiple different needs. And I think we're seeing that value proposition resonate as we engage with more neoclouds.

Yes. I agree. And I'm going to ask you one more open-ended question. I think there are still some details to be worked out, data sovereignty, data security, and I think your 1touch acquisition announced the other night fits into that narrative. But ultimately, I'm looking for an inflection point. We love inflection point where one earning report, you come out with a with a big revenue growth exceeding 20%. Are we -- how should I think about that context? Are we setting up to the -- over the next couple of quarters where you're going to see a scaling of your products? We're finally transitioning into deploying AI among enterprises. Or is this going to be just a slow train moving forward, but it's just not -- is this slow train is not going to turn into a bullet train? Hopefully, that makes sense to your question.

Well, I mean, I would just point out that the revenue guidance and outlook we provided for FY '27 almost doubled our initial outlook coming into growth coming into FY '26. So I would argue that we are driving significant acceleration across both the core business as well as our newer markets, whether that's the hyperscalers, whether it's AI. Look, we're focused on continuing to drive growth, continuing to accelerate. We have multiple levers that we're pulling on. And just to reiterate for your audience, certainly, the core business focused on completing the portfolio -- well, we've completed the portfolio focused on layering on top of that, the intelligent control plane and data management capabilities, things that will allow us to reach into -- and hire into larger accounts, compete for larger wallet share. And you see the results of that already starting to pay bear fruit with -- I think, Tarek shared a number. The number of deals that we track and have booked over $5 million, growing 80% year-over-year that, again, gives you a data point in terms of the leverage we're driving in the core enterprise.

Hyperscalers, we've talked a lot about that in terms of the tremendous opportunity there. And notwithstanding, we're in a bit of a tough position to give too much disclosure because of the customer concentration. But needless to say that business is growing ahead of our expectations, and we've tried to share some color on that. And then certainly, you and I have had a lot of discussion around AI and the number of different angles that Pure -- I guess, Everpure, I should say, is playing in AI, whether that's in the enterprise space with KV cache and inference, whether that's really reaching deeper into the neoclouds with FlashBlade exit. So we've got a number of levers. We're focused on sustainable growth and accelerating that over time more so than individual inflection points.

Yes. And I think your execution over the past several years is a proven point with how well you're situated with enterprises. Perhaps it's the hyperscalers that if I were to use a scenario analysis could provide a growth acceleration. Would you agree?

Well, I think that any of those levers could provide growth acceleration. And we're looking to drive acceleration out of out of all of those. But yes, I mean, just by kind of the size of the market and so on, I mean the hyperscalers, it has dominated our conversations, is a huge focus area for us, and it's one of the big drivers that we're really leaning into.

Do you feel like you're in a situation where you want to drive operating growth profit? Do you want to drive operating margin expansion, but you also have all of these opportunities? So how do you manage the OpEx, right? Because you can allocate more OpEx to drive growth acceleration. But then you do give up some margin. Am I thinking about this right? You have to deal with a delicate balance, right?

Yes. So [ Leah ], let me share kind of our views on where we're focused. I outlined and we discussed all of the exciting kind of areas that we see as capable of driving growth, the levers that we're pulling on to get there. So how do we then think about resource allocation and how do we think about driving growth versus profitability? And the short answer is our thoughts and philosophy on that really haven't shifted.

Charlie, Tarek, myself, the entire management team, we are -- make no mistake, we are focused on driving growth. We're focused on leaning into those levers, but at the same time, continuing to drive moderate leverage over time. Where we've tried to quantify that, just kind of roughly for the Street is we want to drive top line growth while every year accreting roughly speaking, a point to the OM. That said, and I think what Charlie -- what Tarek has articulated as he's joined is the better metric and where we're really more focused on is driving OP or operating profit dollar growth. Now obviously, you can do the math and back into what the implied OM is. But that's kind of how we're thinking about the balance of investments, growth opportunities, but then also continuing to drive incremental improvements to profitability.

Does the 1touch acquisition help you with hyperscaler opportunity, enterprise or both? And that was very interesting. My interpretation of 1touch acquisition is that, yes, it does help you with streamlining the data solvently, data security. So maybe you could just quickly explain what 1touch acquisition is all about? And how is it going to impact enterprise, hyperscalers or both?

Yes. Happy to. So let me hit the first thing -- the second thing first. It's primarily targeted. We see it primarily targeted towards the enterprise and the parts of our portfolio that exists today, but also, we will build in the future targeted at the enterprise customer base. What is 1touch? What is the strategic fit? I talked a little bit about this, or we did on the conference call. But if I step back for a minute, they really provide a couple of critical capabilities that we will both continue to sell in their existing offering, but also integrate into our core technology over time.

These critical capabilities are the ones that come to the top of the list our data discovery, being able to automate data source discovery, data classification, right, being able to look at different pools of data and then generate from that and understanding of, okay, what is this? Is this PII? Is this customer data? Is this health records, et cetera, et cetera? And then the third is the ability to contextualize that, the ability to represent and capture semantic meaning of the data. If you net all of that out and you look at what we have been articulating as our Enterprise Data Cloud vision, our intent is to integrate those components into Purity, into Fusion, into the Enterprise Data Cloud over time so that we can go help customers better understand the meaning of their data across multiple different sources, multiple different silos and work with their data much more seamlessly.

Charlie has talked about enterprise data is still trapped, is still subservient applications. We do a great job of providing the infrastructure for that today, where 1touch and the integration of those capabilities will come in and help is it will help customers actually be able to integrate data across multiple sources without having to do the old school thing of ETL and copying data all over the place. That said, we just announced the intent to acquire, we have yet to close the transaction. We expect that to happen at the beginning of our Q2. And then as I was telling somebody the other day, then the work began. So then we begin with the technology integration. But very excited about it. I think the strategic fit is quite exceptional.

And this goes back to what Charlie referenced earlier in the call, the unified control plane, and I'm not sure how many people picked that up, but when we think about unified control plane and given where the data lakes located in various places, the 1touch helps consolidate and actually make the unified control plane more effective, right?

Yes. I think it complements it very nicely. Think of it as the unified control plane helps customers consolidate infrastructure, where 1touch integration will help is they'll help customers consolidate the logical data and the use of data. So it's very complementary.

Yes. And what's also fascinating, I'm just making an observation here, enterprises wanting to do inferencing or deploy the infrastructure for inferencing on their own, given the data security and everything that you already mentioned. So -- and I do get a sense that maybe Pure is also strategizing to capitalize on this. So how do you deal with the actual deployment of necessary server rack? You are still a point solution, but you offer a comprehensive storage portfolio. Is that your enterprise customer that has to think about deploying the compute and storage together? Or is there like a third party that is involved? Any -- am I thinking about this right? Feel free to correct me what I just laid out.

Yes. So I think the essence of your question is, hey, we're a great data storage infrastructure provider. I'm talking a lot about additional software. How does the customer go and deploy that? We don't ship servers. What does that look like? So look, we work very closely with, I would say, converged infrastructure partners, Cisco being probably the strongest one, where we do have converged -- everything from converged solutions and offerings to strong reference architectures with multiple server vendors. And of course, our strong partner in GSI and systems integrator community works to help with deployment for customers.

What I'd say is that especially with modern applications, the shift to cloud native, really Kubernetes and container-based deployment of applications, in some ways, has made this part of the puzzle a lot easier to fill in. Frankly, the value and where customers are focused is how do I deploy containers and Kubernetes quickly. In most enterprise customers, they're getting to the point where if you can walk in and say, hey, I've got the software offering, it's container and Kubernetes deployable. They've got a Kubernetes team somewhere that knows how to go do that. In some sense, the specific servers and racking and stacking that really has become commodity. And so it's not an area that we're particularly focused on.

And do you have any efforts to work with different CPU architecture? There's also competition on who is actually -- which kind of a compute architecture would do the inferencing, the XPUs and CPU and so forth. It's a diverse set of technologies out there, right?

Yes, there's a diverse set of technologies. We work with -- and really our customers deploy solutions on just about all the major CPU vendors, major and less, I would say, smaller GPU providers as well and everything in between. Clearly, in the GPU space, the bulk majority of our customers, whether it's training or inference, are running on NVIDIA, whether it's NVIDIA GPUs in third-party servers or NVIDIA provided servers. NVIDIA continues to be an incredibly strong partner for us and a focus -- really a focal point of our AI solutions development, as you might imagine.

Sorry, there's a fire drill going on. Apologies for background noise.

A literal fire drill.

Yes. I want to pivot back to the supply chain. And focusing on the enterprise side because on the hyperscalers, they're doing -- they're helping you with procurement. But on the enterprise side, how much of a visibility do you have with the demand? Is there like a backlog? Because you have to turn around and think about procuring the NAND, especially the tightness that may not go away. So help us understand the contracts you have with the NAND suppliers for enterprise application? And how are -- is there any flexibility to adjust the shipment? And how does the pricing kick in here? How often does the pricing is negotiated?

Yes. I mean -- so what I'd say is that we're -- our supply chain team for years, has been top notch. You saw the incredible work that they did to position us well through -- to weather the post-COVID supply chain tightness in calendar '21. The nature -- I'm not going to get into specific contracts and kind of numbers, but the nature of our strategic agreements and long-term agreements with our critical suppliers hasn't changed. If anything, that's really strengthened over time. We, as you might imagine, have multifaceted sets of agreements that, at the end of the day, give us very good visibility into our enterprise supply.

I will say, and Charlie touched on this on the call as well, there are many components that go into a finished solution. The -- I would say the tightness on the overall supply chain situation right now is not just constrained to NAND, as you would know, covering the industry, it stretches everything from NAND to hard disk drives to HBM, high-bandwidth memory to DRAM to CPUs to networking cards. So our teams are very hard at work and busy ensuring that we have the supply we need to meet the enterprise demand that, as I mentioned before, is quite robust. We do benefit from the same structural advantages that I think helped us through the post-COVID tightness.

And for your audience, just as a reminder, because of our software efficiencies, we're able to use far less components to the order of 20% to 30% than a lot of our enterprise competition. The fact that we have one unified hardware technology means that our overall BOM or bill of materials is a lot lower. There's fewer different parts to go chase. I have a lot more flexibility in steering mix in terms of a lot of shared parts means that for whatever I source, I have more flexibility to build all the things in my portfolio. And so these are things that I think we're also leaning into.

Got it. Two quick follow-ups. Can you update us on the mix of the QLC NAND that you're using? And how does that mix changes into the second half of the year?

So yes, I mean, we've been shipping more QLC than TLC as a business in the enterprise, so setting aside the hyperscalers for quite some time. I think that mix continues to shift more and more towards QLC, but it's been over 50-50 for quite some time. I haven't looked specifically first half, second half, but the mid- to long-term trend is pretty clear. It's kind of the QLC portion is growing.

Yes. And to what extent are you able to pass on the incremental cost increase to your enterprise customers?

So I think -- and the best way to look at that is the commentary we gave on gross margin implications, which is that -- which is to say we do expect near-term pressure in Q1 gross margins on the product side, specifically because the input costs grew very, very rapidly in the back half of Q4 faster than the ability for the market pricing to adjust. We've since taken pricing action. We do expect the effects of that pricing action to really help us recover those gross margins starting in Q2 and then throughout the year. That -- all of that said, we don't -- to be very clear and to restate, we don't we don't think of our pricing philosophy and our pricing actions as being gross margin driven.

We look to the market. We price in a competitive space. We command a premium for our solutions. That's reflected in our long-term gross margin benefits over the competition. What happened here was that, again, the rate of input cost increases was almost unprecedented faster than the market could adjust pricing. And so what we've done in those pricing actions is we've reevaluated, reassessed where market pricing has moved to accommodate the new higher level of input costs. We've adjusted our pricing, and that's really what you're going to see starting to flow through Q2, Q3, Q4 so that the gross margins recover back into a healthy part of that long-term range of where we want to be in the product side, 65 to 70 points.

Great. Okay. So we're a minute over the time. I want to thank you for giving us opportunity to catch up with you post earnings. I want to thank everyone on this call for participating. If there's any follow-up from investors, feel free to send me an e-mail, [email protected]. I wish everyone a great weekend. Thank you, Rob, and the rest of the IR team.

Thank you.

Thanks, Mehdi.

Bye.

Good afternoon, everyone. I think we're about to get started. I think it's the last event of the conference, I believe. So anyway, thanks for joining, everyone. Just a reminder, I'm Mark Newman, Bernstein's U.S. IT hardware analyst, and great pleasure to welcome Charles Giancarlo, CEO of Pure Storage.

Charlie, please, my friend.

Charlie, Charlie. Charlie. So before we get started, I have to read -- I'm told I have to read the safe harbor. So statements made in these discussions, which are not statements of historical fact are forward-looking statements based upon current expectations. Actual results could differ materially from those projected due to a number of factors, including those referenced in Pure Storage's most recent SEC filings on Forms 10-Q, 10-K and 8-K.

So now we've got past that. I'd like to perhaps, Charlie, if you could start off, you just reported earnings and you also I believe you just changed your name to Everpure. Maybe if you wouldn't mind just summarize why the name change and maybe just briefly summarize your earnings before we jump into the questions, if that's okay.

Yes, absolutely. Monday morning of this week, we announced our name change as well as an acquisition, which I'll spend a minute on. And then just yesterday, of course, we announced our latest quarterly is our fourth quarter, so last quarter and last year's full earnings release.

So Monday was an interesting day. We changed our name. We've been Pure Storage for about 16 years. And so the question might be why change the name? Well, we've really very much redefined the enterprise storage industry in many different ways. We've reached a point to where while we started out with a niche products fitting into an important but small percentage of a corporation storage needs.

Today, we can really fulfill any storage need that -- and every storage need that our customers have. So it really makes us a full supplier across their needs. And being the -- not just being the newest, but being the company that invests the most in R&D in this area, we do it with a single operating environment with a very consistent hardware architecture. And we do it with the world's highest Net Promoter Score and highest customer satisfaction, the highest reliability.

So that's a fundamental change in where we sit today. As of today, not only do we spend the most from the standpoint of percentage of revenue and R&D, we actually invest more R&D in data storage and management than any one of our competitors in the business, including the largest companies. So it's a big change in terms of our position.

Secondly, we have, over the last several years, been investing more and more in data management. That is making the data more manageable and more useful to companies, especially as they look to analytics and AI. On Monday, we also acquired a company called 1touch that accelerates our activity in this level to where we can add context to the data we store so that data becomes more meaningful from an information standpoint for input into analytics and AI. So that's really, again, accelerating our movement into this data management space.

So why the name change? Well, we were Pure Storage and the personas that make decisions around data management, around data security tend to be different people than the people -- than the storage admins who make decisions around storage. And storage in our name was actually holding us back from having conversations with those other people because those other personas would say, well, I don't manage storage. I don't need to speak to these people. we have other companies that we speak with. So it was a very simple name change.

We're still pure. We're just more pure than ever. Therefore, Everpure, and it's going to open up the aperture for some of the conversations. Now a year from now, is it going to matter? No, we're not a consumer company, but this was about opening up the opportunity to speak to more personas and making that easier for our sellers.

So yesterday, we announced earnings. Just to give you a snapshot, it was our first $1 billion quarter. So we were roughly $1.60 billion. Last quarter, we announced that was roughly 16% year-over-year growth for the full year. It was actually 20% growth quarter on -- that is Q4 this year versus Q4 last year. And we also announced that -- or we guided Q1 to be a 28% growth quarter year-over-year and for the year as a whole, a 19% year-over-year. And that's compared to last year, where we guided 11%, we delivered roughly 16%. And next year, we're guiding 19%.

And one more comparison point, we had a couple of our competitors announced today. They're in the low single digits in terms of growth rate. So our continuation of our core business to gain share is, in our view, accelerating at this point.

Would you like to briefly comment on the acquisition before we jump into?

So yes, the acquisition of 1touch. 1touch is a company that builds a product that is able to search for data repositories within customers, and then it is able to identify the data within those repositories. It's able to classify it, understand the context of the data, look for data, understand the security posture of that data, determine what's at risk, whether there's, for example, PII, personally identifiable information in that, able to mask it and provide a variety of other services that we plan on adding to our -- adding that context to our data within our storage environments so that, again, that data now is not just raw data, it's data that's been contextualized that has really effectively gone through an ETL chain in operation, in its operational form to be able to provide a richer data set to analytics and AI.

Great. Appreciate that. So if I just kicking off some questions, and then we can see if any questions from the audience. I'd like to talk a bit about NAND flash given what's going on in the market with high pricing and shortages. So how do you view Everpure -- I have to get used to saying your name. How do you view Everpure's positioning within the NAND supply chain given the supply constraints and dynamics at the moment?

Yes. Well, we're on a path because we -- as of last year, for the first time, are selling not just to enterprise customers, but to hyperscalers as well. So our responsibility for NAND flash this year will be multiple times what it was the year before, okay? So we do have long-term contracts in place. But the spot market for NAND has gone -- like memory, like CPUs has gone crazy. The rate of change of pricing in the spot market, the world has not really seen anything like this ever.

Now unfortunately, given the events of 2021, 2022, the whole tariff craziness last year and now this, unfortunately, we're using unprecedented too many times in the same decade. But really, this is remarkable. Some prices on some products have more than doubled in less than 4 months. That's to give you a sense of the rate of change of these things.

So it is -- and the other thing is there are a lot of shortages now that are popping up in the market because there's just more demand than there is supply. So it's going to be a topsy-turvy supply market for probably the remainder of this year. We've weathered this very well in the '21, '22 time frame. We think we'll weather this as well as anyone, if not better, this time around, but there is a bit of unpredictability to all of this.

Are you NAND, are you constrained -- are you constrained by your supply of NAND flash at the moment?

Not at the moment. No.

Not at the moment. And with regards to the NAND pricing, does your pricing framework allow you to immediately requote existing backlog? How does that impact your margins? Is there a timing difference between your cost rate change?

Rate of change makes a difference because when we provide pricing to customers, well, generally, your teams, as you might imagine, they're talking to customers quite a bit before a proposal goes out. So expectations are set. Then a proposal goes out pricing. That proposal usually has a time frame on it. Typically, in the past, it was 90 days. It's now been shortened to 60 days, and it may shorten even further. Many of our competitors now are not only issuing pricing that's only valid for a week, but some of them have said, actually, we reserve the right to change pricing until the day we ship. So that's sort of the span of what you're seeing in the market. So when costs change faster, then you can change pricing in the market, that's when gross margins get a little wonky.

So there's going to be a bit of a transitory period. Price, as costs go up, it takes time to reflect it?

That's correct. Once you reach some cost and price stability, then we don't expect any change in gross margin.

Once the NAND price stabilizes, then you believe those costs will pass through. So your margins will be...

Absolutely. It's a competitive. Just to be clear, we're not competing against our price last year. We're not competing against raw commodity prices. We're competing against other system vendors who have the same challenges on input costs that we do.

So next topic, I'd like to talk about NAND flash versus hard disk drives. Everpure recently published a white paper suggesting that by 2028, virtually no new all HDD enterprise data center systems will be sold. But data suggests at the moment that hard disk drives still represent around 80% of gigabytes in data centers.

In hyperscalers, not in enterprise.

Right, right. Not in enterprise. So maybe you could just help us, help reconcile that view.

Well, the current supply-demand market, which basically has exhausted supply of any type of data storage. In fact, hard disks are sold out through '27. Flash is effectively sold out through '27. So customers are just buying whatever they can that satisfies their needs. So that transition has slowed down a bit. But these cost curves of both semiconductor, believe it or not, you can tell I'm not that young. When I was in business school, we studied the cost decline of magnetic hard disk. That was quite a long time ago.

Innovators' dilemma.

Yes, Innovators' -- well, it was pre-innovator's dilemma, I hate to say, but yes. And the fact of the matter is the hard disk, if you extract any particular year and just look at it on a long-term basis, has followed that cost curve for 50 years, okay? Except for any individual year supply-demand imbalances, the cost decline of flash has followed its -- you could call it Moore's Law, it hasn't been around as long, but for many decades. So the current price or supply-demand imbalance is a temporary phenomenon, hopefully temporary. We will get back to those curves. There's always a reversion to the mean. I may be off a year or 2, but I'm not off on a long-term basis.

So I guess my question is, you're talking about all hard disk drive becoming less. Basically, you're calling for flash to take more share longer term in storage in the data center. However, if you talk to some of the hard disk drive makers and the NAND makers directly, they're saying that they're not really changing market share. They're kind of going after separate markets. I mean how do you -- I mean how do you think about that? I mean how do you think about -- do you think that 80/20 is going to change? Do you think that 80-20...

I have literally no doubt whatsoever. First of all, let's -- why is it that we are able to sell into the hyperscaler market and none of our competitors sell into the hyperscale market. We have a very unique technology, right? Hyperscalers today use 1 of 2 different commodities for their storage. They use either hard disk or they use SSD. We don't provide hard disk, we don't provide SSD. We provide something we call direct flash. And most people say, well, wait a minute, SSD is flash. Not quite. SSD is flash masquerading as a hard drive. Literally, it is. That's why it worked in your computer automatically and why you can switch one to the other.

But flash is a semiconductor, a magnetic hard drive is a mechanical device. They work very differently. So to take a semiconductor, which is a very powerful technology and make it look like a mechanical device, there's a whole bunch of translation and other logic. There's a little computer inside an SSD that is translating what flash looks like, makes it look like a mechanical hard drive. We don't do that.

We use raw flash. Our software allows that raw flash to be used the way that flash wants to be used, and we connect that to the application. We're literally the only company in the world that does this. Okay, so you might say, great, what's the benefit? The benefit is we get a 30% to 40% price performance improvement by doing that versus SSDs. SSDs waste price performance. Now so why do people use them? Because disk is managed out of the operating system. Nobody likes to change operating system environments. And so SSDs were a brilliant way of getting flash into a computer very rapidly.

All of our competitors were based on hard disk. They could go to flash very quickly by substituting SSD. Guess what? Hyperscalers started out on hard disk to be able to accelerate performance, they could throw in SSDs, easy-peasy. Now that storage is 25% or more of their space power and cooling and cost, getting an extra 30%, 40% out of that. We are 1/10 the power space and cooling of hard disk. We are probably twice the power or 1/2 of the power space and cooling of SSD, and we're more performant.

So there are a lot of reasons why hyperscalers will use our technology. And that goes even beyond this, I won't get too techy on you, but that will benefit here. And to answer part of your question, what is the benefit over hard disk, Well, 1/10 space power and cooling, when 25% of your space power and cooling is storage is a big deal, and we are same total cost of ownership. So...

So total cost -- basically TCO is lower even though the dollar per...

Even if price higher, TCO is lower.

Right. Actually, there was a lot of discussion last year around -- I think it was September, maybe August, September about NAND. It was actually SSDs. So maybe that's a little bit outside of what you exactly do, but SSDs taking share from hard disk drives due to the shortage of hard disk. Did you hear -- are you aware of that? Is that...

I'm not only aware of it but I can confirm that, that is, in fact, taking place. As I said, there's a shortage of everything. There's not enough of everything to go around. And so there's more than one reason why even SSDs in hyperscaler, which if they're not using us -- well, even if they are using us, they're also using SSDs. But if they're not using us, there are only other choices SSDs. And there are several reasons, one of which being not enough hard disk another of which is hard disks don't have the performance that these enhanced data centers need and therefore, they're going to more SSDs.

But given that now -- I mean, back then, NAND wasn't as short and pricing wasn't really going up that much. This is August, September. Now that NAND is also probably even more short than hard disk drives and pricing is going up dramatically, is that going to shift back to hard disk drives?

No. I mean hard disk drives are also short. So there's -- at this point, they're just desperate for anything they can put in place.

Okay. So you've argued before that for AI hyperscalers, the real bottlenecks are data center shell and power costs, not drive costs. So there's 2 interesting dynamics playing out right now; power is increasing, increasing becoming a limiting factor, which clearly favors NAND, but offsetting. But then on the other hand, you have this rapidly increasing NAND prices versus hard disk drive prices per terabyte. So how do you think about that?

Hard disk drive prices are going up now rapidly as well.

What kind of -- how much are you seeing?

Over 50%.

Per terabyte?

Yes.

Okay. Well, the recent numbers from the hard disk drive makers is implying up like 2% year-on-year terabyte so far reported. So that blended average, though. So if there's some specific contracts that are up 50%, that's...

Spot pricing is going up as much as 50%.

I see. Well, still NAND pricing has, in some case, spot prices up like $5.

Double.

It depends on which contract. Anyway, depending on which contract you look at, NAND price is clearly up much, much more than hard disk drive. So there's 2 competing dynamics there. Just maybe you could just help us reconcile that and talk about TCO and how that plays into this.

I will. But I think we're spending too much time on -- I mean, on really basic hardware and not enough on software and services and especially with respect to our business is being driven by the enterprise business, to be very clear, okay? And we're spending a little too much time, I think, in my opinion, relative to our revenue and growth on this particular topic. But that being said...

I'll change the topic next question.

Okay. So I'm sorry, I forgot the question.

No, I was saying the 2 competing dynamics of power is becoming more of a bottleneck. NAND is better for power, but on the other hand, NAND price going up more.

Look, we're in a topsyturvy world in terms of lack of supply, so shortages plus pricing. Right now, the hyperscalers seem to be -- demand seems to be price insensitive. And what I mean by that is they want to build out their data centers. They are desperate for delivering the data centers on time at the capacity that they need. They seem to be price insensitive. That's not true in the enterprise, but it is very true in this current race that we have. And so it's really less about almost everything else pricing than it is about just getting the bits. And I'd say that's what's driving most of what's going on right now.

Okay. Yes, changing topics a little bit. So -- your company is, I think, the first storage vendor certified for NVIDIA's DGX SuperPOD, making it like a true Day 1 choice for organizations building industrial scale AI factories. So direct flash, your product remains a key differentiator for your company. So I just wanted to ask you to assess your competitive positioning in light of the new Vera Rubin platform and what Jensen Huang talked about at CES, given how much more important storage is going to be and then how it's becoming rearchitected in the Vera Rubin platform?

Yes. Well, it's interesting. This is -- so another new segment of the storage business because it is a new segment. It's not the -- at best, it will become 10% of the overall storage market. So I want to put it in perspective. Is the AI segment, specialized storage for AI, of which we have a product that -- now one of the things that is -- we use the direct flash, but another thing that makes our system unique is we use the same software and the same hardware architecture for all storage needs, block file an object, large-scale, small scale, high performance, low cost, all the different protocols, including the very, very high-performance AI environment, right, which is what we use for SuperPOD, for interfacing with non-hyperscale GPU-type environments.

Hyperscalers, we use our hyperscale architecture. But in the case of GPU clouds or any of the tech titans that are building out large-scale environments, we use what we call our FlashBlade//EXA platform. That platform provides the world's highest performance. And in the case of some of the newest elements associated with this type of system, such as KV cache, which is a key value cache. We are able to deploy literally the best benchmarks in the world in this area. So it's a relatively new market, but as I mentioned, it really is very specialized. It goes to the GPU clouds generally in the tech titans.

That's not the problem that exists in the enterprise environment. In the enterprise environment, most of whom will not be actually building their own GPU environments. They'll be using, in many cases, the GPU clouds. The enterprise challenge is that their data is not yet ready to be used by AI. I think for those of you that have been in the data management market, you know that in order for data to be prepared to be used for analytics or AI, it has to go what's commonly known as an ETL chain, extract, transform, load.

This ETL chain is held together with many different bits of software that come from generally different companies and a lot of people, a lot of human middleware, a lot of data scientists that put this together, usually weeks, if not months of work to transform data that's very raw coming out of operational systems like databases. And add all the context and meaning and semantics to it, so it becomes, I'll call it, information to sort of distinguish it from data, becomes information that can be used for AI, okay?

That information is stored in yet other systems, usually other storage systems, many of which we provide and other compute environments. So we think of this as being very wasteful in many ways. One is it's wasteful of human labor. Two is wasteful of time. You're not using real-time data. You're using data that's been transformed over weeks or months. And AI, you'll see this often, your AI isn't giving you timely information, certainly not in a business environment. It's giving information that's based on information that's weeks, if not months old.

For business, they want to be able to operate on their real-time information. So where we're going, which is part of this acquisition that we announced, is being able to add the proper context, the proper information semantics to the data as it's being created in the first place so that it's more ready to be used by AI. When you talk about, for example, RAG, retrieval augmented generation, that's a term that was created by NVIDIA, what they're doing is they're saying, okay, well, we're training on old data, and we're coming up with an old response, but then we'll go out and check with the real-time data to see if it makes sense. We're trying to bypass that entire gap between real-time data and data that's already been transformed or data that takes weeks to transform. So that's really where we're focused.

Got it. Got it. Fantastic. That's very helpful. I believe you've surpassed your FY '26 shipment target of 1 to 2 exabytes.

Correct. This is to the hyperscalers again.

But you've also described FY '27 as mixed and nuanced revenue model, I believe. I think we're talking about hyperscaler versus enterprise. So I just wanted to understand, given the current engagement you have with one big hyperscaler. What does that means essentially for your margins and revenue going forward?

That's a huge question. Okay. Let me see if I can hit it piece by piece. So we had announced last year that our expectation for this year was double-digit exabytes into the hyperscale part of our business. that we are standing by that. In fact, we now believe it will be larger than what we had previously identified. But we didn't give a specific number on it. We've become very cautious. We've become very careful now with our public statements as to the exact nature of what we're shipping and to whom because we do have a primary hyperscaler, and they have told us they don't want their proprietary information being shared. So we have to be more thoughtful in how we describe these things.

But in any case, this year that is -- let me go by calendar years. This 2026 calendar, we will ship a lot more than we shipped in 2025, and we believe we will ship a lot more in '27 than we shipped in '26. And so that's that. We've refined and now standardized our -- the economic model by which I mean what does the gross margin picture look like in the operating margin picture of the products that we will ship to the hyperscale customer base.

And last year, it looked more like it was going to be nearly all software. This year and the standard that we've created for a wide variety of reasons, which if we really want to get into, we could do that. But suffice it to say that as we look at it this year, we're going to have gross margin in that business between about 75% and 85%, and that's our standard model as we go forward.

Okay. So before it was more -- it was mostly software for hyperscalers. But now going forward, it's going to be blend of software and?

It's a blend of some amount of hardware. The answer to that is our direct flash modules, which are not SSDs. But remember, when they buy an SSD, they buy one thing, it's one price. They don't have to worry about how it's built, where it's built. and so forth. The previous model required that they would build the DFMs. The DFMs have lots of different -- have mostly flash, but they have other components on there. It put the burden on them.

They don't want that burden. We're taking that burden. So while they are still -- we are not procuring the NAND, that will be procured separately, but we are procuring all the other components. So that requires a slightly lower gross margin. It's not providing us any less margin overall. It's just that now that those components will flow through our P&L rather than someone else.

And given your exposure to enterprise, are you seeing much of adoption of AI data centers, AI servers in enterprise versus in the cloud?

Very little. It's going primarily into the same enterprise customers that were previously buying what's called HPC or high-performance computing environments. So no, it's largely going into the GPU clouds, obviously. It's going into pharma. It's going into automotive. It's going into quant funds, but they were buying GPUs and "supercomputers" already. So it's going -- all national labs, obviously. And what is different is that sovereign clouds are more and more buying GPUs, and they were not a factor before.

I see. And are you seeing like significant attach for storage when they implement any kind of local on-prem AI service?

Yes. But I want to put it within reason. Again, I'll say that the storage that's going to be associated directly with AI is still going to be about 10% of the storage market. There's a lot more hype around it in the store, which is very different than networking and obviously compute, very different. In, let's say, a traditional data center, storage is about 25% of the overall capital purchase, if you will. In the AI data center, it's less than 10%. It's just there's -- compared to the cost of GPUs and the cost of 400 gigabyte -- gigabit rather, 800 gigabit network switches and so forth, it's just a smaller component.

Did you have any exposure to neoclouds?

A lot, yes.

And so one of the things I found quite interesting from one of the other guest speakers we got here, I think it was one of the hard disk drive makers was that neoclouds, they actually -- they're building their own GPU servers, AI servers, but they are actually using storage at the hyperscalers. So they don't have as much storage locally. Do you see that?

So there's a lot of that. And let me describe why. There's multiple tiers of storage. It's only the very hot tier of storage that is the very, very fast tier of storage that they really need to have directly connected to their GPU clouds. All of the other storage is effectively data in waiting and the performance requirements on that can be quite a bit less. And so for some of them, if some of them want to outsource it to the major cloud providers, there's no reason not to. I think part of the reason why they do that is because they're all cash strapped. And so the less that they need to spend on CapEx, the more that they can spend in an OpEx way. If you're a start-up growing with capital needs, the less you need to spend on CapEx, the better.

I see. But you're still seeing a need for flash storage, direct flash storage locally for the neoclouds.

On that high-performance tier.

I see. Got it. To what extent is...

By the way, we see it both on the high-performance tier and the lower performance tiers. Not all of them are going to the cloud for the lower performance tiers.

I see. I see. Okay. Can you talk a little bit about your Nutanix partnership? Is that acting as a pull-through for FlashArray sales in the enterprise? Can you just talk a little bit about how that relationship works and what's the impact?

Yes. No, I'd like to expand that. So we're seeing a really big demand now for alternatives to VMware, to be very direct. After the Broadcom acquisition, in many cases, the price for VMware has been increased by a factor of 5 or more. And many customers, even if they've signed on to a new license, are looking to get off of VMware in the next few years. There are a few alternatives for them. Nutanix is one, and that's been a very good partnership with Nutanix.

Another one is Kubernetes virtualization. So moving to what's colloquially called KubeVirt for Kubernetes virtualization, generally running on either SUSE or Red Hat. And our product, which is called Portworx, which operates both for -- it provides for staple storage for both container-based workloads as well as Kubernetes virtualization workloads. It's almost the only solution. It's practically required for any scaled Kubernetes virtualization project.

We're seeing huge growth in that area. Customers are desperate to move away from VMware in many cases. Their time frames are limited by their current contract. They want to get -- they want to move their environment by the end of that contract, and it's driving a lot of demand for both of our solutions.

Okay. I would like to see -- open up to the audience to see if there's any questions from the audience. Any questions? We've got a microphone here. That's all super clear. Do you see? There's one question here.

I just want to share a little bit more practically how it actually works. So I work for Fidelity. We need storage.

Thank you. You're a wonderful shareholder, a very long-term shareholder.

But how would you approach a customer like that? Just how does it work? How do we purchase things from you? Or do we use -- how do we make decisions on where to store our stuff?

Yes. So we didn't really spend a lot of time on our core business, which is growing at unbelievable rates and where we're picking up market share like crazy. We boast the world's highest user Promoter Score, right, Net Promoter Score at 84. That was our latest for last year was 84. Our competitors are in the 30s. Now why is that? Why is our Net Promoter Score so high? Our product literally delivers -- and this is not an exaggeration. In fact, it's an underestimation. Our product delivers more than 10x greater reliability. It requires 10x fewer people to manage, right? Generally half the space power and cooling.

So we have a lot of bona fides to our name. We basically give all of the storage admins out there, their weekends back because they don't have to worry about us failing, which is a major problem in data centers is storage failures, right? So we provide a remarkable -- and we're the only vendor that provides nondisruptive upgrades forever. So we have a unique program called Evergreen. What is Evergreen? This is unique in all the industry. When our customers buy our product, and they take what we call our subscription.

So this is -- don't think of this as a service contract. It's a subscription. What does that mean? What it means is we consistently upgrade and update the product every year, year after year on a consistent basis so that after -- now our earliest customers are about 12, 13 years old. If you were to visit that one purchase that they made and visit that installation, that product looks like a product we sold last year.

In other words, the product is always new. It'd be like buying a car once and always having a new product without having to go through a disruptive upgrade, without having to turn your applications off, remove it, put it in a new one. No, no. This is upgraded on the fly without taking the applications down. It sounds like magic, and we can take you through it.

We're the only ones in any industry that does this, nondisruptive upgrades, just like you expect from a SaaS platform, okay? When you subscribe to a SaaS service, you don't expect it to go down. You do expect them to upgrade it. You expect them to improve it every year, but you don't expect any downtime. That's what we deliver on-prem, all right? Now we also delivered it entirely as a service. You don't even have to buy it day 1. You just pay for what you consume, and we deliver that as a service. So that's another unique offering.

So part of the reason why we're growing now well into the double digits, our competitors are flat. maybe growing single digits is because of this incredible service that we're able to provide. Now we're the only ones that deliver -- so storage is typically subservient to the application. Each array is managed on its own to that application, right? We, over the last year, have delivered a new upgrade to our software so that all of the customer's storage appears as a cloud.

So they might have tens or hundreds of arrays across their worldwide environment. It just appears as a cloud service. And what they do then is set up policies for how data is handled. And the service, if you will, takes care -- the software takes care of where the data is placed, how it's managed, what policies are set against it. So they're managing their data rather than managing their individual storage.

So our typical gross margin is in the 65% to 70% range. Our competitors' gross margin is in the 40s. What's the difference? The difference is software. The difference is we're managing the customers' data. We're not just providing them a storage system.

Great. Thanks very much. Any more questions from the audience? I don't see any hands. Maybe Charlie, to wrap up, I mean, what do you think people are missing here for your company, Everpure? I mean, is there something -- maybe just kind of just to wrap up, things you think Wall Street are missing, things that you want to add that you think people don't understand about your company would be?

Well, I think our long-term shareholders who we've had for 10 years now aren't missing anything at all. They know our company very well. We've had the same long-term shareholders honestly, they are fantastic supporters of the company. They understand our strategy. They understand where they're going. The hyperscale business of ours has, in my view, unfortunately attracted a lot of hedge funds that are betting, okay, this quarter, they will announce a new hyperscaler, run up the stock and then we don't announce a new hyperscaler and they run out of the stock.

So we have a lot of volatility now, which I would much prefer not to have. But the hyperscale part of the business is something that we're betting on. It's going to be a good long-term investment. In the meantime, our I believe we've hit in a sense, escape velocity on our core business, which is after all, we're roughly -- we're almost $4 billion going over -- hopefully going over $4 billion this year in a $40 billion business. We have a lot of market share to make up there.

And that's a very exciting part of our business, and I think we're hitting escape velocity. We have the ability now to overcome all of our competitors in that space. And I think that there's too much focus, if you will, on the -- not that it's not exciting, but when it's less than 10% of our revenues and it extracts 60% of the time on analyst calls, it's just too much focus on that part of the business rather than the core, which is growing really well.

Got it. Okay. Well, great. Charlie, thanks very much for joining us today. We appreciate it.

Yes. Thank you.

Thanks so much. All right. Cheers.

Good day, and welcome to Everpure's Fourth Quarter Fiscal 2026 Financial Results Conference Call. Today's conference is being recorded. [Operator Instructions] At this time, I'd like to turn the call over to Paul Ziots, Vice President of Investor Relations. Please go ahead.

Thank you. Good afternoon, everyone, and welcome to Everpure's Fourth Quarter Fiscal Year 2026 Earnings Conference Call.

On the call, we have Charlie Giancarlo, Chief Executive Officer; Tarek Robbiati, Chief Financial Officer; and Rob Lee, Chief Technology and Growth Officer. Following Charlie's and Tarek's prepared remarks, we will take questions.

Our press release was issued after close of market and is posted on our website where this call is being simultaneously webcast. The slides that accompany this webcast can be downloaded at investor.purestorage.com.

On this call today, we will make forward-looking statements, which are subject to various risks and uncertainties. These include statements regarding our financial outlook and operations, our strategy, technology and its advantages, our current and new product offerings, supply chain, hyperscaler opportunity and competitive industry and economic trends.

Any forward-looking statements that we make are based on facts and assumptions as of today, and we undertake no obligation to update them. Our actual results may differ materially from the results forecasted, and the reported results should not be considered as an indication of future performance.

Discussion of some of the risks and uncertainties relating to our business is contained in our filings with the SEC, and we refer you to these public filings.

During this call, all financial metrics and associated growth rates are non-GAAP measures other than revenue, remaining performance obligations or RPO, and cash and investments. Reconciliations to the most directly comparable GAAP measures are provided in our earnings press release and slides. This call is being broadcast live on the Everpure Investor Relations website and is being recorded for playback purposes. An archive of the webcast will be available on the IR website and is the property of Everpure.

Our first quarter fiscal year 2027 quiet period begins at the close of business, Friday, April 17, 2026.

With that, I'll turn it over to Charlie.

Thank you, Paul. Good afternoon, everyone, and welcome to Everpure's Q4 and Fiscal 2026 Earnings Call. Q4 was an outstanding quarter. Our first billion-dollar revenue quarter capped off a strong performance for FY '26, with full year revenue of $3.7 billion. And we enter FY '27 with strong momentum.

Revenue in Q4 was driven by broad-based strength across our business, particularly in Enterprise. We are executing a clear strategy to modernize and simplify data infrastructure for our enterprise and hyperscaler customers amid rising AI demand, power constraints and increasing operational complexity.

Our Enterprise Data Cloud architecture continues to strongly resonate with customers, with over 600 customers adopting Fusion since its introduction a year ago.

Consistent with our core strategy of investing in data storage as high technology rather than a commodity, Everpure can now support practically all enterprise storage needs and use cases with our unified Purity operating environment and our Evergreen hardware platform.

Purity, enhanced with Fusion, now adds a unified control plane to its unified data plane, and enables customers to manage their global data as their own enterprise data cloud with consistent enterprise policies implemented in software. The focused investment in our enterprise business is translating into accelerating demand and growth.

Our Purity software, DirectFlash architecture, and Evergreen promise, have proven their flexibility and universality by extending smoothly into our FlashBlade//EXA offering. As a reminder, FlashBlade//EXA supports AI-scale workloads at industry-leading performance and efficiency. FlashBlade//EXA has achieved industry-leading MLPerf benchmark performance and recently published the highest results in the SPECstorage AI Image benchmark. This past quarter, in a strong competitive context, we secured our first EXA customer and are in advanced stage discussions with dozens more.

Our hyperscale business grew beyond our expectations in FY '26. We have broadened and expanded our solution and we've standardized on our financial structure, which Tarek will detail later.

Entering FY '27, we expect continued growth in our hyperscale solutions concentrated in the second half of our year, which is also incorporated in our guidance. This broad and increasing momentum is reflected in our strong FY '27 revenue guidance of almost 19% year-over-year growth at the midpoint which Tarek will also discuss in more detail.

As I stated earlier, Everpure has now reached a point where we can support the full spectrum of our customers' data storage needs, from high performance to low cost, from tens of terabytes to tens of exabytes, from AI to backup and all protocols, including block, file and object. We support all customer use cases including all databases, all virtualization, containers, file systems, object systems and Kubernetes. We deliver all of this through a single software operating environment combined with our DirectFlash technology and our unique Evergreen architecture and business model. Today, they ensure our customers always have access to the latest technology. This set of capabilities has been developed because Everpure alone has invested in data storage and management and high technology rather than as a commodity. We now invest more R&D in data storage and management than any other competitors.

Our continuing investment in innovation is increasingly focused on enabling our customers to better control and make use of their data for AI and analytics.

We released our latest audited Net Promoter Score, which is the gold standard for customer loyalty and satisfaction. We achieved an industry high NPS of 84 for calendar 2025. While much of the industry remains with scores in the 30s, we have maintained a score above 80 for more than a decade, while increasing our customer base to over 14,500. This is yet one more reason why we continue to outperform all other competitors in the market.

We have developed a strong brand that is well known for quality, consistent innovation and strong customer care. Our position has strengthened to where we are now gaining franchise customers who put their trust in us to standardize their IT architecture on our platform for a majority of their storage infrastructure.

We are now in a position to not only help our customers automate their data storage, but increasingly to enable them to better manage their global enterprise data.

With Fusion and our Enterprise Data Cloud architecture, customers can apply policy-driven governance across workloads, moving beyond traditional storage to standardize, protect and intelligently manage their data sets. This represents a meaningful evolution from infrastructure management to comprehensive data governance.

Earlier this week, we announced a definitive agreement to acquire 1touch, which will accelerate our ability to help customers unlock the strategic value of their data and make it ready for AI.

1touch's technology delivers discovery, classification, governance, cyber resilience, data sovereignty and context to prepare data for AI and serve as a critical foundation for the Enterprise Data Cloud and enterprise-scale AI deployment.

Our rebranding and name change reflect our growth from operational storage to intelligent data management, empowering customers to extract greater value from their data in an increasingly AI-driven world.

Our new name Everpure, reflects both what we have created and where we are going. We are expanding our brand to align with our expanded horizon and to attract a much broader set of strategic personas.

We are entering FY '27 with strong momentum, and we expect continued growth across all 4 products and sectors. These include commercial, enterprise, government and hyperscaler, our U.S. and international theaters and our systems, software products and services such as Evergreen subscriptions.

Turning to the broader environment, we expect macroeconomic uncertainty to persist through the coming year. Strong component demand, driven by tech titan AI buildouts as outstripped supply across the industry, dramatically increasing NAND, memory and CPU pricing. We expect that the industry, including Everpure, will see unpredictable component shortages, which could lead to extended lead times and potential shipment delays.

As we have identified in previous calls, we have highly distributed and resilient supply chain and have weathered past supply chain disruptions well.

Supply chain constraints are operating as both a tailwind and a headwind in our hyperscale discussions. A bit of a tailwind as hyperscalers are more eager to accelerate their testing and certification of new sources of supply and as a headwind to every vendor's ability to source the necessary components.

Everpure raised prices on our product line on February 9, reflecting the dramatic and rapid rise in component prices. I believe we were the last in our industry to raise prices and I also believe that our increase was the lowest in the industry to protect our customers. Based on the extraordinary rapid rise in component costs, we expect product gross margins in Q1 to be at the lower end of our typical range of 65% to 70%, but we also expect them to recover through the fiscal year.

We have built a diversified supply chain with contingency plans to reduce disruption risk even as the industry faces shortages. Our long-standing and direct component supplier relationships and in-house hardware design provides us additional flexibility in addressing supply chain disruption.

Our Evergreen model provides transparent pricing that protects customer economics. In addition, with our continuous improvements in performance and capabilities entitled in Evergreen, existing customers will benefit from our new data reduction software. This recent release of Purity Enhanced Data Reduction offsets our higher pricing by providing a cost per effective terabyte that is lower than our previous prices on some workloads.

We are at a very exciting time in the story of our company, Everpure. We can now compete for all of our customer storage infrastructure. We provide the world's most consistent, comprehensive and reliable data storage environment. Our Enterprise Data Cloud architecture will allow customers to more efficiently manage their global data, and we are now creating the technologies that will allow our customers to more easily prepare their data for their AI future.

Our focus on investing in data storage and management as high technology is driving the accelerated growth that we see today.

With that, I'll invite Tarek to provide further details.

Thank you, Charlie. We closed the year on a high note, with Q4 revenue surpassing $1 billion for the first time, representing 20% year-over-year growth and record operating profit of $226 million, implying a strong operating margin of 21.3%.

Our performance in Q4 was broad-based, with particular strength in the enterprise. We increased the number of customers transacting during the quarter and delivered solid performance in large-scale transactions with deals over $5 million growing 80% year-over-year. This performance was supplemented by accelerated growth in the government sector, along with several notable Enterprise Data Cloud platform wins.

We also secured the first sales of FlashBlade//EXA, signaling positive initial market interest and demand for this new offering designed for large-scale artificial intelligence and high-performance computing workloads.

Moreover, we expanded our footprint with our existing hyperscale customer, delivering strong growth in our hyperscale business for the year, ahead of our initial expectations. We are confident in the sustained momentum of our hyperscale business. More on hyperscalers and our year-end performance later.

But before we get into that, earlier this week, we made 2 key announcements for the company, including our intent to acquire 1touch, a leader in AI-driven contextual data intelligence.

1touch delivers top-down, automatically discovered and enriched contextual view of data across the data center, cloud and edge. This critical software capability enables our customers to better understand the meaning of their data and unlock its strategic value through AI and other applications. In turn, the acquisition of 1touch will further differentiate Everpure by allowing us to embed unique data management capabilities into our core Purity software offerings.

Founded in late 2017, 1touch has a financial profile of a fast-growing company that is investing heavily to gain traction in the market. Unsurprisingly, it is not yet profitable. We expect 1touch to be 1.5% dilutive to operating profit in fiscal year '27 and to become accretive to operating profit within 24 months from the acquisition on a post-synergy basis.

As a company, we are expanding beyond being a storage provider to becoming a comprehensive data infrastructure and data intelligence platform. We are transitioning from simply delivering storage solutions to redefining data management at a global scale. Our new name, Everpure, reflects this transformation and captures our new identity as a full-scale data intelligence company.

To conclude, we are uniquely positioned to address the full spectrum of our customers' needs and to compete for large strategic enterprise franchise opportunities, supported by our newly expanded AI-enabling platform.

Now let's deep dive into the details of our fiscal year '26 performance and subsequently discuss our outlook for fiscal year '27. Q4 product revenue of $618 million grew 25% year-over-year, while fiscal year '26 product revenue of $1.97 billion grew 16% year-over-year.

As a reminder, our product revenue category now includes revenues that we receive from hyperscale shipments as well as a portion of Portworx software revenue when sold as term licenses.

Q4 subscription revenue of $440 million grew 14% year-over-year, while fiscal year '26 subscription revenue of $1.69 billion grew 15% year-over-year.

Q4 Total Contract Value sales for our Storage-as-a-Service offerings grew 28% year-over-year to $179 million, driven by high-velocity transactions of less than $5 million. For fiscal year '26, TCV sales grew 32%, totaling $520 million for the year. This significant year-over-year growth momentum reflects the increasing adoption by our customers of Evergreen//One and other subscription-based offerings, which deliver a consistent, nondisruptive operating and management environment.

In fiscal year '26, total revenue grew 16% to $3.7 billion. We also delivered our highest annual operating profit of $635 million, an implied operating margin of 17.3%.

Turning to gross margins. In Q4, total gross margin was 71.4%, supported by a robust subscription services margin of 77%, while product gross margin stood at 67.3%, an increase of over 400 basis points year-over-year, driven by a favorable product mix.

It is important to note that sequentially, our product gross margin in Q4 was lower than in Q3 '26 as we had lower hyperscale shipments and Portworx license shipments in Q4 relative to Q3.

I would like to remind everyone that these sales are lumpy in nature. In addition, the sequential change in product gross margin reflects changes in customer and product mix during the quarter. The variance also includes a modest impact in the quarter from increasing component costs, which prompted pricing actions taken early February '26.

For fiscal year '26, total gross margin was 72.1%, an increase from 71.8% in fiscal year '25.

As industry-wide AI-driven infrastructure demand continues to outpace supply, driving higher input costs, we expect continued component price volatility across the storage industry as well as extended lead times and potential shipment delays. As a result and as Charlie mentioned, we implemented price increases across our product portfolio on February 9, 2026.

It is important to remember that while we maintain long-standing supply agreements with our NAND suppliers, these agreements help mitigate but do not eliminate significant input cost swings and potential shortages.

As we mentioned in the past, historically, component cost volatility has had a greater impact on our top line and on margins. When component costs such as NAND rise, the industry typically sees higher pricing as competitors face similar input cost pressures. This dynamic supports improved pricing discipline and can act as a tailwind to revenue over the medium term with some short-term pressure on gross margin as prices catch up to extraordinarily rapid cost increases.

We remain committed to treating our customers fairly and will not engage in price gouging or take undue advantage of the current market dynamics.

Moving on to our subscription business. Q4 subscription services revenue of $440 million increased 14% year-over-year, accounting for 42% of total revenue. ARR grew 16% to $1.9 billion.

I am particularly pleased with the results of our remaining performance obligations, or RPO, which accelerated to 40% growth in Q4, driven by the execution of large deals and strength of our Evergreen//Forever and Evergreen//One offerings. Notably, our RPO pertaining to our subscription services offerings grew 34% exiting Q4.

Turning on to revenue by geography. In Q4, U.S. revenue grew 9% to $674 million, while international revenue increased 48% year-over-year to $385 million. International revenue represented 36% of total revenue. The continued expansion of our international footprint remains a significant opportunity and a key strategic focus for the company.

For fiscal year '26, U.S. revenue grew 12% and international revenue increased 25%. In fiscal year '26, we expanded our customer base by more than 1,100 new customers, including 335 in Q4 alone, reflecting continued momentum throughout the year. Our penetration of the Fortune 500 now stands at 64%.

With respect to our organization, in Q4, our head count increased sequentially by 166 employees, bringing our total head count to 6,400 employees at year-end.

Our balance sheet remains robust with over $1.5 billion in cash and investments at the end of the year. Cash flow from operations in Q4 was $268 million and $880 million for the year.

Capital expenditures during the year were $264 million, representing approximately 7.2% of revenue for fiscal year '26. Our capital investments during the year supported data center expansion, the increased testing of new products and solutions, the scaling of our hyperscale business as well as the funding of initiatives aimed at accelerating Evergreen//One subscription growth.

Free cash flow for Q4 was $201 million and $616 million for the year. Free cash flow margin for the year was 16.8%, tracking our operating profit margins of 17.3%.

In Q4, we repurchased 1.7 million shares, returning approximately $127 million to shareholders. For fiscal year '26, share repurchases totaled $343 million or 5.6 million shares. For fiscal year '26, 56% of free cash flow was utilized for stock repurchases.

In addition, we paid $68 million in withholding taxes on employee awards in Q4, offsetting dilution by approximately 1 million shares and $271 million for fiscal year '26, offsetting about 4 million shares.

We currently have about $329 million remaining under our existing $400 million repurchase authorization announced in Q4 '26.

Now turning to our guidance for fiscal year '27. As Charlie remarked earlier, unprecedented component demand driven by AI buildouts has outstripped supply across the industry. At this stage, the duration of the demand-supply imbalance and related risks to the industry are hard to predict. Needless to say, we are actively working with our suppliers to mitigate these risks and navigate this period of uncertainty.

For Q1, we anticipate revenue to be in the range of $990 million to $1.01 billion, representing approximately a 28% increase year-over-year at the midpoint. We expect operating profit to be in the range of $125 million to $135 million, representing approximately a 57% increase year-over-year at the midpoint.

For fiscal year '27, we anticipate revenue to be in the range of $4.3 billion to $4.4 billion, representing an 18.8% year-over-year increase at the midpoint. We expect operating profit to be in the range of $780 million to $820 million, representing approximately a 26% year-over-year increase at the midpoint.

In terms of seasonality, we're entering fiscal year '27 with very strong momentum and expect 47% of our revenues to be generated in H1 of fiscal year '27, which represents a 2% improvement year-on-year.

Let me finish by adding more color about the factors that underpin our guidance for fiscal year '27. First and foremost, and specifically for our hyperscaler business, I would like to remind everyone that we started ramping our hyperscaler line of business in fiscal year '26 and feel very confident about the future of our business for years to come. We expect to significantly accelerate shipments and revenues in fiscal year '27 relative to fiscal year '26. This momentum is reflected in our strong fiscal year '27 guidance. As a reminder, hyperscaler revenues are governed by the schedule of hyperscalers data center buildouts and are not linear during the course of the year.

For fiscal year '27, we expect the majority of revenue from hyperscalers to be recognized in Q3 and Q4. Also, we have now standardized our business model to cater for the hyperscaler market. Moving forward, we will procure some of the components that are needed by hyperscalers to build their solution in their environment, but not the NAND. Hyperscalers will continue to procure the NAND through their supply chain. As a result, we expect gross margins of hyperscaler revenues to range between 75% to 85%, a level accretive to product revenue gross margins and overall company gross margins.

For Q1, we expect product revenue gross margins, excluding hyperscaler gross margins to be at the lower end of our typical range of 65% to 70%, reflecting the impact of unprecedented and sudden increases in NAND and other component pricing. As we move through the fiscal year, we anticipate gross product margins will recover.

Second, and in line with prior commentary, we will continue to invest in R&D and sales and marketing to fuel growth in our core and establish our brand and these investments are factored into our operating profit guidance.

In terms of our fiscal year '27 operating profit guidance, and as mentioned earlier, we are absorbing a 1.5% dilution to our operating profit for the year from the acquisition of 1touch. Yet we continue to be laser-focused on accelerating growth and building operating leverage as our guidance attests.

With that, I'll now turn the call back to Paul for Q&A.

Thanks, Tarek. [Operator Instructions]Operator, let's get started.

[Operator Instructions] Our first question comes from Amit Daryanani from Evercore ISI.

I guess maybe my question really is around the fact that you folks have had some really impressive results right now, given all the fixation everyone has on memory.

Charlie, perhaps you can just help put in context the acceleration in revenue growth we're seeing in Q1 and really fiscal '27. How much of that is really coming from your ability to pass on the memory price increases to your customers versus perhaps you're just seeing a step-up in demand, especially if you think about customers at this point, perhaps need to find a way to improve the utilization rate of flash, which seems to be going up in price a lot. It seems like Pure is uniquely positioned to help provide a solution here. So just on kind of what's driving the growth and just flesh out how much of it is surprising versus customer step-up in demand?

Yes. Absolutely. Well, I can certainly assure everyone that Q4 in particular and Q1 is really all demand-based because, first of all, we took no pricing action in Q4. Furthermore, most of our proposals and bids that go out, go out with the 90-day pricing.

In other words, we put out a proposal, customer has 90 days to accept or not. So those price -- a lot of that -- of Q1 shipments will be based on Q4 wins, of course, and that will be based on our old pricing.

So most of this new pricing that we put out earlier this month won't really hit until our Q2. So therefore, both our guidance, if you will, for Q1 as well as our print for Q4, really all demand-based. And that's one of the things that gets us very excited as we go forward, of course.

Our next question comes from Aaron Rakers from Wells Fargo.

I guess I wanted to maybe double-click on the breadth of what you're seeing in hyperscale opportunities. I know you talked about more of a weighted towards the second half as far as a ramp. But as we think about the changes you're making in procurement, the gross margin expectations, I'm curious if we could read into that at all of kind of the opportunities maybe expanding into additional hyperscalers through this year or any progression you've had with the engagements with other opportunities?

Yes. Look, we are seeing increased activity. And as I sort of alluded to in my opening remarks, the level of activity is going up -- well, the amount of activity and the breadth of the activity, both are going up. Now we have said before and we'll say again that until we have another one to announce, we're still in the fight and not yet at the finish line. And you might say, well, it's a bit later than we expected. And I would say that's true.

What I would say is that advancement continues and becomes broader as we go along. A lot of this is very dependent on the hyperscalers' own testing and development plans. But what we're seeing is wider interest, broader engagement and wider engagement.

Our next question comes from Howard Ma from Guggenheim Securities.

Great. And congratulations on the strong financial performance and also the rebranding efforts. So either for Charlie or for Tarek, when I look into the qualitative commentary about higher pricing serving as a potential revenue tailwind, the momentum in EXA sales and maybe even initial shipments to a second hyperscaler as Charlie you just alluded to, it would suggest that you exit FY '27 on a high note. But when I look at the full year revenue guidance, I believe it implies that growth would decelerate to 13% to 14% in the back half. And I understand the back half comps are tougher, but you also are talking to higher hyperscaler shipments in the back half.

So my question is, does the guidance factor in accelerated enterprise demand in Q1 that could come at the expense of future quarters? Or are you intentionally haircutting your assumptions, pipeline close rates more than normal to account for the uncertain environment?

Howard, thanks for the question. It's Tarek. I would say to you, we ended up fiscal year '26 and Q4 in particular, with a very, very strong performance.

Linearity in Q4 was very much back-end ended in the quarter, which as a result, leads us to have a very strong Q1 guidance. And if you really look at the difference in growth rates between the Q4 attained growth rates of 20% and what we're guiding at about 26% at the midpoint for Q1, that shows you that we have a very strong momentum carried from fiscal year '26 to fiscal year '27.

And so the rest of the fiscal year '27 is in line with our internal expectations. And this is why in my script I highlighted that seasonality H1 versus H2 revenue seasonality is more weighted towards H1 this year as a result of the strong finish in fiscal year '26.

Our next question comes from Mike Cikos from Needham & Company.

I just wanted to cycle back. So first, congratulations on the EXA customer win that you had cited. I know that you also discussed how you're in advanced stages with dozens more at this point.

Can you just walk us through what that EXA customer win actually looks like, how that sales cycle unfolded? What were some of the key points around that win? And then I know it's an earlier offering, but where are we in establishing, I don't know if you want to call it your customer testimonials or playbook to make this a repeatable process on your side?

Yes. Thanks for the question, Mike. This is Rob. I'll take that one. Yes. So really happy to see the initial customer win actually multiple sales of EXA in the quarter. As you mentioned, it's a newer offering, which we went generally available with, I believe, at the end of Q2 or beginning of Q3. And we've seen really since launching the offering at last year's GTC conference in March, really strong initial demand.

With this particular GPU cloud customer, the nature of the win and how we got there, frankly, this customer was looking to stand up storage to provide for their end customers' training and inference workload environments. They had actually gone down the path of initially selecting an alternate vendor. They engaged with us, did a performance test, were frankly blown away with the performance and turned around and placed an order that within a couple of days. That subsequently led to follow-on orders as they were able to deploy that, get it into production quickly and really light up their end customer footprint.

As Charlie mentioned in his prepared remarks, we're in various stages and some advanced stage discussions with dozens of other customers. If I step back from it, I would say that EXA really is filling a market gap and a market need of covering both off on the high performance that's truly demanded of these environments. But also at the same time, providing for the reliability, the usability, all of the manageability aspects and simplicity that Everpure is known for.

Our next question comes from Samik Chatterjee from JPMorgan.

My question was on the product gross margin. And you have maybe more on what's driving the confidence in guiding to the trough margins being in Q4 into Q1 and then steadily through the year. How much of the underlying commodity that you need for your full year guide do you have visibility into already in terms of locking in through either LTAs or contracts with your suppliers? And are you able to lock in pricing to the extent that's driving the confidence. Would want to understand sort of the underlying sort of drivers there as much as possible.

Yes. Samik, thank you for the question. It's a -- let me try to make the answer as straightforward as possible. Our gross margins tend to be relatively steady with respect to cost changes in the supply chain. That holds true when cost changes are relatively gradual. The cost changes that have taken place literally over the last 4 months have been anything but gradual. They were very rapid.

And what that does is it breaks the synchronization between our pricing, which generally has, as I said, about 3 months or more time in the market and our ability to fulfil that pricing with costs because the costs have very dramatically increased in a shorter period of time.

So we believe that we can get back to our standard gross margins as the pricing stabilizes. Stabilizes doesn't necessarily mean stay steady, but it means that it doesn't change by the significant factors that has changed over the last several months.

Visibility is a very different thing. I think visibility is just non-existent. The prices have been changing so rapidly, and the market is so dynamic, whether it's visibility or contracts, frankly, at this point in time, there's very little.

Our next question comes from James Fish from Piper Sandler.

I appreciate the time here. I'm sorry, jumping between calls. Just trying to understand how much of a pass-through the flash you have in sort of your BOM here? Are you passing on in that price increase versus eating? And what was the price increase magnitude that you guys actually instilled on February 9 there across the business, understanding flash as a percentage of the business can range pretty drastically across each of the products.

Yes. Well, first of all, flash -- it's not just flash, all the components, memory is very well publicized, but even GPUs. Those components are both in short supply and prices are very dynamic, I would say, right now. On average, they've more than doubled over the last 6 months. So think about that. That is a very big change.

We think the pricing will address the increases that we've seen so far. That pricing increase was on average -- about average 20%. Now when I say average, it's very different on different product lines for the reasons that you mentioned, different amounts of componentry costs associated with each different size NAND. But frankly, a lot of the controllers, of course, have a lot of memory in it as well.

So it's a complicated formula, but on average, about 20%. And as I stated, I believe actually, first of all, we were, I think, the last to announce a price increase. And from all the price increases that I've heard from our competitors, I believe ours was the lowest.

Our next question comes from Simon Leopold from Raymond James.

I appreciate the breadcrumbs you've given us on the royalty business here, some of the shifts. So it sounds like the messages are around lumpiness and a back half load to the current fiscal year. But what I didn't hear you update was the prior commentary about expecting double-digit exabytes this fiscal year. And I appreciate double-digit could be 10 to 99. So can you put a finer point? My sense is you feel better about the outlook, but I want to check in.

Right. So a previous commentary quarters ago were that we expected low double digits in this year. And I can confirm that will probably be the same, but more than what we expected, more than the amount that we had expected when we gave that commentary last year.

Our next question comes from Wamsi Mohan from Bank of America.

I appreciate all the color around pricing and memory. I guess, Charlie, would love to get your perspective on what you're seeing from a customer behavior standpoint? Given you just mentioned that the environment for supply is actually quite unpredictable.

Do you think that you saw any pull forward of demand? I mean you're also saying first half is going to be stronger. Your pricing doesn't really get impacted as much in the first quarter, clearly, maybe even not so much in second quarter. So to some degree, like the pricing is yet to happen, but you're seeing a significant uptick in demand. And second half, you're expecting hyperscaler revenues, including hyperscaler revenues to still be more -- less weighted than typical in second half.

So just curious if the purchasing behavior that you're seeing in your customer conversations are indicating any pull forward or any level of maybe demand destruction for the industry, maybe not for you, but demand destruction for the industry given some of these unprecedented price changes.

Well, I think we're seeing several things. One is, we are seeing increased demand for our product. Win rates, size of the wins are increasing. So we believe we continue to pick up share in a meaningful way.

As I mentioned in my script, what were -- what might be called franchise deals, and I'll explain that in a second, those are going up. And what we mean by a franchise deal, the customer is not talking to us about or the conversations that are not just about this workload or this new opportunity, but pure, let's have an enterprise scale discussion about you becoming our 1 of 2 or even our sole storage strategic partner, those are going up, right? So the quality of the conversations are increasing.

And of course, when you do that, the pricing is not a transactional pricing. It really is about building a relationship and a structure that works well for both us and for the customer overall.

So we're seeing demand increase I'm going to leave the quantification, if you will, of pull-ins, but what I would say is that customers certainly are -- they have a lot of projects that they need us for.

Generally, when prices -- when costs went down and prices went down, they bought more, but not enough to offset the price decrease. We expect the inverse now. We expect there to be, in fact, some elasticity reduction, that is negative elasticity, but probably not enough to overcome the price increases that are occurring across the industry. So we do expect it to -- generally, the market dynamics to be such that it increases somewhat the total market. And then...

Yes. Let me elaborate on what Charlie has said, Wamsi. It's Tarek. First of all, I want to reemphasize the fact that we have had a number of very large deals that we booked in Q4. Like I said in my remarks, deals of value that is in excess of $5 million increased 80% year-over-year, which is a very good result from my perspective.

We finished Q4 extremely strong. Linearity was towards the end of the quarter. And the timing of it has to do a lot with the frenzy of price increases that the entire industry had to make to adjust for the sudden and rapid rise in underlying input costs.

And so that's, in itself, yes, you're correct, there is an element of pull-ins. But the amount of pull-ins, I would say, is not a substantial amount that underpins our Q4 performance. It more underpins, per my answer in the prior question earlier, the growth rate in Q1. And if you want to quantify that, it's probably a mid-single digit in the growth rate that we have guided for Q1 fiscal year '27.

Finally, I would say that we're seeing very strong demand in our Evergreen//One offers. And if you really look at our performance measured in terms of RPO increase, our RPO has increased 40% in Q4, which really a test of the validity of the model and the fact that in uncertain times, customers tend to turn to solutions like Evergreen//One for their needs.

Our next question comes from Erik Woodring from Morgan Stanley.

Tarek, I just -- given the comments you guys were making to Samik earlier just about kind of volatile pricing, uncertain supply, elongated lead times. And then I guess, potentially the risk of any demand destruction depending on what your customers do or how the industry is forced to raise prices in the second half. Can you just help us better understand kind of the philosophy or the approach you took behind setting your full year guidance, just given all of these relative unknowns, especially as we look into the second half of the year?

Sure. So as always, our guide is determined by the level of demand and deals that we see in the pipeline. So our guide reflects the strength of the pipeline that we have and visibility of the demand that we have looking forward. We feel very good about the visibility of that demand for the next couple of quarters. Beyond that point, it's hard to ascertain.

And we also have factored into our guidance, obviously, the price increases that we have made. Those price increases were timed well, I would say, because we effected them on February 9, which gives us pretty much the benefit for the full year. And we went very surgically around adjusting our prices, knowing that we have had a pretty strong benefit coming from Enhanced Data Reduction, so that we make sure that whatever price increases we put forward still provides value to our customers.

And thirdly, of course, our guidance factors in the benefits from our hyperscaler business, which is going to be accounting for a lot of the growth in fiscal year '27. And like I said in my opening remarks, the hyperscaler business revenue is not linear during the fiscal year. It tends to be governed by the schedule of the data center buildouts. And in this case, we expect the vast majority of the hyperscaler revenue to come in Q3 and Q4 of fiscal year '27.

Our next question comes from Param Singh from Oppenheimer.

It's interesting to me to see how you're moving up the software stack, particularly with the acquisition of 1touch.

Maybe you can talk more about your thought process here and what else may be required from either an organic or inorganic investment to increase the value proposition for customers, particularly for AI workloads.

You bet. So -- and some of this is outlined a bit in our presentation that has come along with the earnings announcement. The way we look at it is the following. Customers today, actually, generally, they manage their applications and their applications manage the data that's associated with that application.

The customers and many customers don't actually manage the data itself. And that -- if they do, it's all with people, human middleware. What we've created with Fusion and the Enterprise Data Cloud is the opportunity for data sets that is the same data that's used in different places at different times for different purposes to be managed as data. That is to be managed on a global basis, to be managed by policies that customers set in software and then the data is managed as such.

So instead of managing storage, they're managing data. What we're doing with 1touch now is we will be able to add the ability to provide context around the data. When -- there is the shorthand that we speak about a when we talk about AI using data, doesn't really use data. It uses data that's been highly transformed into information that is self-describing so that AI can use it. That process of going from data to self-describing information is done with a lot of work by companies.

Companies have to use extensive ETL tools. They have to use a lot of human labor to restructure the data so that it can be used by AI. We believe that a lot of this work can be done in the operational environment that is with additional capability that we build adding not just to our customers' data, whether it's on our product in the SaaS environment, on the cloud or even on third-party product and add that context to the data where it lives, and really just make it a lot easier for AI to be able to use data on a real-time basis. And Rob, do you want to add to that?

Yes. Just to add to what Charlie said. I think we -- as Charlie mentioned, we've been expanding from our roots in data infrastructure through to data management, bringing the cloud operating model into the enterprise data center with Fusion and so on and so forth. This next leg of our enterprise data cloud vision really is strongly supported by 1touch and the strategic fit is quite exceptional.

1touch is bringing in critical capabilities to go and advance that vision road map, capabilities in the areas of data discovery, data classification, really being able to understand the semantic meaning of data that Charlie -- as Charlie was just outlining.

In short, if I step back from it and I look at our portfolio evolution and where we have grown from our roots in data infrastructure to Fusion to Enterprise Data Cloud over the last couple of years and where we're headed.

I would say that with Fusion, we really started making data infrastructure AI-ready, really being able to connect all the different silos in the data center and then this next leg with 1touch. Integrating that technology is really going to go serve to help customers make their data itself secure an AI ready.

Our next question comes from Krish Sankar from TD Cowen.

Yes, I have a question for Tarek. Very impressive growth in RPO sequentially. And I'm just wondering, are hyperscale revenues included in that? And how much of the $700 million of sequential growth is from hyperscalers?

Well, the answer, Krish, is none because hyperscalers revenue is recognized in product revenue, it's not part of -- it's not ratable. And we do not include it in the remaining performance obligation.

Our next question comes from Asiya Merchant from Citigroup.

Good job on the name change as well. Just -- you talked a little bit about the hyperscaler opportunity, low double-digit exabytes that was talked about earlier. Just if you can give us any update on now that you have a more structured model in how you're approaching these hyperscaler deals, if there's any update on the progress beyond the first hyperscaler announcement?

And related to that, the investments that you're making exploring those opportunities, what specific -- if you could drill a little bit deeper into what sort of investments are those? What are you looking to expand? Are those more related to product or with the nature of those investments, that would be great.

Thank you, Asiya. Of course, this is a difficult area for us to chat about for lots of different reasons. But what I would say is that we are in engineering test environments in multiples of the hyperscalers now, and they're proceeding well. But it still takes more time before we know whether or not we are certified or whether there is a specific date by which they plan to use us on buildouts. So until we can get to that level of certainty, we just have to leave it there, really.

And then on the nature of the investments, I wouldn't call out any change in what we've described before, primarily looking at continuing to help our hyperscaler customers grow and scale, R&D to go support that as well as scaling the supply chain to support that.

Our next question comes from Tim Long from Barclays.

Yes, maybe for Tarek, just wanted to touch on something on the subscription services side of the business. You guys saw a really nice rebound in the TCV business in fiscal '26, a nice acceleration. At the same time, we're kind of seeing a deceleration on just the overall subscription services revenue number compared to the prior year.

Just curious, is this just timing? Is this related to maybe the impact of the non-TCV business where product sales were lighter in '24 and '25. Or is there something else going on where we're not seeing those 2 kinds of metrics tracking a little bit more closely together and kind of how do we think about that going forward?

Yes. Thanks, Tim, for the question. I think you need to look at subscription services revenue growth in conjunction with the RPO growth. It's really important to look at them together.

The RPO growth is really the best measure of latent revenue and momentum that we have in the business.

What is interesting is that this is the first time where we are lapping the introduction of Evergreen, and we are renewing a lot of those Evergreen contracts, which is really good.

Some of those renewals are longer term than what we originally anticipated. It's good for us to drive longer-term renewals. It's the test of the validity of the model. And that elongation of the renewals is what goes into the RPO. And this is why the ARR, for example, is growing at 16% and the RPO is growing at 40%.

And that explains why you're asking your question and you're looking at it as a deceleration from FY '25 to FY '26 in terms of growth rates. And you're right, it's 22% in '25 and 15% in '26. But again, the reason is we are shifting to longer-term contracts. That is the main reason. And the RPO, a test of the performance of our subscription business.

Looks like we have time for one more question. So the next question will be the last question.

Our last question comes from Max Michaelis from Lake Street Capital Markets.

You guys shifted away from primarily operational storage and more into the data management space. I mean how does this rework your competition landscape? I mean previous partnerships may now be bumping up against them in terms of competition. So I guess if you can just kind of give us an idea on how that changes the competition landscape for you guys.

Sure. Thanks, Max. I want to be clear, we're not shifting away from storage. We are definitely not shifting away from storage. We are adding -- yes, we are absolutely adding data management. So we actually believe, in some ways, this is a new area. Now it's certainly true that this space was filled by ETL software and systems and a lot of human labor in our customer base. But the entire ETL segment is being disrupted by AI as it is.

So that's an area, not only in disruption, but an area that's ripe, I think, in terms of opportunity, of instead of completely separating out systems of operation, operational systems, which is what a lot of our data storage gets sold into and then systems of information, which is other storage and other compute that is completely dedicated to actually analytics and/or AI. And these are just 2 completely different environments.

You might ask, well, why replicate all of your data away from the operational system into a system of information that has to go through a lot of ETL and a lot of transformation before it can be used. Why not have the data that's sitting in the repository of -- in the operational repository, why not have that be more ready to be used by AI, in which case, it's much more real time, it's much more valuable and it can be managed along with the foundational data itself on a global basis nearby policy rather than I've mentioned in the past, by human middleware and fingers on keyboards.

So we think this is a new opportunity. It's a new architecture, if you will, for how data is managed in the enterprise, and we think it really sets us up in a unique place.

And Max, this is Rob. Just to add one thing to what Charlie said, which I think was the other part of your question, looking to understand how does this fit into our partnership landscape.

As we mentioned before, 1touch provides critical capabilities that we see longer-term applications for, as Charlie outlined, to really broaden and enhance the ways that customers are making their data AI-ready.

That said, these capabilities are today offered as a DSPM platform. DSPM or data security posture management is an important tool. It helps customers make their data more secure. But at the same time, we rely on a strong ecosystem of partners to provide a much more holistic set of security capabilities and fully intend to keep working with those partners on our customers' behalf.

Before we conclude, I think Charlie has some final comments.

Yes. Thank you, Paul. And I want to thank all of you for joining our call today. We know there were many other calls out there, some, very large. We're entering a new chapter in our company's journey, and I'm proud of the progress that we've made and the momentum that we're seeing right now.

I want to thank our customers, our investors, our employees, our partners, our suppliers for their trust and partnership, and we look forward to building on this momentum this coming year. Thank you all.

That concludes the Everpure Fourth Quarter Fiscal 2026 Financial Results Conference Call. Thank you for your participation. You may now disconnect your lines.

Great. Thank you, everyone, for joining Needham's Growth Conference here. My name is Mike Cikos, I'm the lead analyst covering cybersecurity and infrastructure software. I'm pleased to say that we have with us Coz and Sandra from Pure Storage for this upcoming fireside chat.

I have a list of prepared questions on my side. We'll be going through but we also have a Q&A function for you guys as well. [Operator Instructions] I'm just going to go through some forward-looking statement disclosures and then we can jump into it.

Statements made in these discussions, which are not statements of historical fact are forward-looking statements based upon current expectations. Actual results could differ materially from those projected due to a number of factors including those referenced in Pure Storage's most recent SEC filings on Forms 10-Q, 10-K and 8-K.

With that out of the way, thank you again for the time, guys, and Coz really appreciate it. Maybe just a dust off for the audience here. But could you just walk us through your background as well as a quick overview of Pure Storage for anyone who might be new to the name of revisiting the story here?

Sure. Thanks for inviting me, by the way, Mike. So I'm John Colgrove. Everybody knows me as Coz. I'm the founder of Pure.

I've been in the storage industry a long time prior to Pure. I have worked for many years at VERITAS. I was one of the first engineers in VERITAS and worked on the VERITAS Storage Foundation products. And before that, I had worked at Amdahl mainframes doing storage stuff for them and at Bell Labs when I was growing up in New Jersey. So I've been around the storage industry quite a while.

Pure, we started in 2009 in October. And we started with a fairly simple premise that disc could ceased to improve at the same rate. Flash meanwhile, driven by a whole ton of consumer devices and such, had achieved an economic scale where it was far faster and more adaptable and flexible, and it was time for flash to start replacing hard drives.

We built the product or the company on sort of three simple things: One, leading the industry transition to all-flash. Two is building very simple and easy-to-use products. Enterprise tech was not known as being simple and easy to use. And we've really pioneered and pushed that. And when you think about easy-to-use products, one of the hallmarks of them as well as they are safer, more reliable, people make fewer mistakes. It's just a better way to build things. And you see that in every consumer device that everybody wants to buy. Enterprise tech should follow that. And then the third was our Evergreen business model. Evergreen is a fairly simple concept. When you purchase one of our products with Evergreen forever. We will refresh every component as it ages, non-disruptively plug-and-play hardware swaps, push button software upgrades. And so the idea is that the product just continues to get better and better and better and stays fresh and new all the time so that you never have to rebuy it.

In effect, you could argue it turns capital purchases after some years into subscription purchases as people continue on with it. And we do have of the first 10 arrays we ever sold a whole bunch of them are still in service through the magic of Evergreen. It's not that they look like 13-year-old arrays, they look like brand-new arrays and those customers have never had to rebuy they've just expanded.

And then we follow that on now in our Evergreen/One Storage as a Service, which is even better, you don't buy the array in the first place, you just purchase a successful storage service from the beginning, and we have a set of SLAs and such.

We went public 10 years ago now. And we're still continuing to build products and build the best products in the industry and move forward. So that's a brief history of Pure.

Excellent. And just jumping into Evergreen/One for a second. Let's talk about that. Where is the market today in terms of willingness or desire to adopt? And the reason I ask is, obviously, you punched on a number of items there. You have the SLAs, you have the idea where instead of being on this CapEx cycle you're shifting it to more of a subscription OpEx type purchase. There's multiple benefits there, but it feels like, in some ways, the market itself might need to shift as far as it's thinking about storage. So can you walk us through where the market is today in terms of Evergreen/One.

Well, so there are a set of people who, I guess, they view capital as very cheap. They're just used to buying capital purchases. And so there's a set of people that's definitely buy CapEx and think they want to continue to buy CapEx. We believe that buying the Evergreen/One service is better, but we'll continue to offer customers the choice.

Now I think as prices continue to go up on NAND as it becomes harder and harder to obtain new gear. I think that tends to move people more towards Evergreen/One. So we saw that before a few years ago, right?

Depending on interest rates, financial outlook, et cetera, the demand between CapEx and Evergreen/One purchases may shift. And the other thing I would say is people shouldn't regard Evergreen/One as a an OpEx way of purchasing the same thing. That is not what it is. When you buy with Evergreen/One, you get a set of SLAs where we're guaranteeing the availability of the performance but you also get SLAs around things like ransomware protection. If you suffer a ransomware attack, we will roll in new gear for you to recover on. You have guarantees around your power rack space and all sorts of other things.

It is a better and different product because you are buying set of success, guaranteed performance, guaranteed availability and all of the other SLAs that go with it. And so people shouldn't just say, "Oh, hey, the cost of capital is higher now. I'm going to go with a subscription". They need to recognize the subscription is better. It's a better product to buy. But I do think as financial things vary back and forth, it does increase the demand for the subscription.

And I should have started if we take a step back and think about the overall market. Maybe you can provide an update as far as the market's view or adoption of flash over hard disk drives or solid state drives. How are customers going through that economic decision from a procurement standpoint.

Well, so again, the key to flash is looking at the total cost of ownership, not looking at the buying the actual bits. Now -- and again, you have to remember all the time, hard drive might be cheaper on the actual bits. But you put that hard drive into a storage system and the extra sheet metal, the extra power, the extra administration that it takes because it is more fragile and there are more hard drives and things like that. All costs extra. And so when you look at the long term -- and the flash lasts longer, we guarantee our DirectFlash modules for 10 years. And so you sort of get twice the lifetime you get with a hard drive or even more. And so when you look at that total cost of ownership, it's far more equal.

Now that said, prices of NAND have been going up a lot recently due to the demand shortages. Prices of hard drives have gone up a little, not to the same extent as NAND, but hard drives are also sold out. You can't get them. And so those factors play in as well.

Right now, I would expect that there are fewer people in the market looking and saying, "I'm going to replace my hard drives with NAND" than there were a year ago. And I think when prices swing back, again, you'll see a big uptake in that. But right now, as prices on NAND stay higher, you probably just see fewer people thinking that way because people tend not to think differently. They think the traditional way they always have.

And on that NAND element, too, I know you've spoken about it a couple of times already here. Can you talk about how Pure Storage has historically navigated movements in NAND prices and the knock-on effect to your business model?

So generally, what's happened in the past, and we've been through three, pretty significant cycles of NAND prices increasing a lot and also NAND prices dropping a lot.

When prices go up, people do pay more dollars per gigabyte. They buy fewer gigabytes. But the net result is when prices go up, generally speaking, in the past, that has been a slight tailwind to our revenue. Margin stays pretty consistent because, again, the price people pay per gigabyte goes up, but the higher prices have generally acted as a slight tailwind to our revenue.

And actually, similarly, when prices are like falling through the floor, people buy a lot more, but they pay a lot less per gigabyte. And that's actually been a little bit more of a headwind for us. And it's not some gigantic overriding factor. Think of it as it will increase or decrease the growth by a few points usually. And we've seen all our competitors take prices up. That's good because we have a better product in the market competitors usually do their best by going to ridiculously low prices against us. And when they take prices up, that means they're not going to do that as much.

One of the things, just cycling back to the commentary on hard disk drives and the spinning the heat. I think about that in the context of this AI era, the energy crunch that we're seeing across the broader industry. Is that increasingly point when you guys are going out and selling. I know you've always had this TCO angle, we are able to talk about energy consumption in the data center. But is that an additional tailwind? Do you find that, that is resonating to a degree that maybe it has not historically?

I'd say that's still more sort of organization by organization. There are a set of companies that really buy through their purchasing department who just ignores all that stuff because they're not gold on it. They're gold on what's their purchase price. And in -- and you do see, again, the worries about energy, energy supply, energy cost rises and falls a little bit. Over time, right now, I'd say that's down a little bit. It will probably swing around again.

Europe has decided that they're slightly more afraid of missing out on AI than the energy issues. So they've backed off just a little bit overall on the energy focus, but -- and their power prices are just way higher than here in the U.S. So it is an important thing to them.

But again, the AI is driving data centers and power consumption through the roof. And so in the long run, that's -- again, it's a trend that's really in our favor. And like everything, it's not a straight line. It varies slightly from year-to-year.

And you guys are also building out, I know a big theme for this year at Accelerator the Analyst Day was the data cloud. And so I'm thinking about the stack that you guys have built out -- but maybe you can talk about how this platform is positioned to benefit customers from an AI standpoint. Where is the company investing to accrue additional value on that front?

Yes. So there's a couple of things there. So the enterprise data cloud, which is a major strategic thrust and it will continue to be for several years.

The goal of the enterprise data cloud, if you think about when somebody contracts with a cloud-based service, they operate differently. So when you're buying virtual machines from Amazon or whatever you don't think at all about the physical infrastructure. And you have a very service-oriented thing, "Oh, hey, I've got my credit card here. I can just buy five more VMs when I need it, and I'm kind of -- it's very flexible and adaptable". That's what draws people to the cloud is that flexibility and adaptability.

So with the enterprise data cloud, what we're doing, we're starting with our own products. We do intend eventually to bring other products in there. But we're allowing people to create a fleet and run all of the Pure products as simply and easily as if they're a single cloud service. And so you -- instead of going and provisioning storage and every time deciding where you're going to provision it and things, you provision it by having some people who have defined service level offerings that you will offer to your internal customers.

And then they say, okay, I need 10 terabytes of Tier 1 storage or I need 10 terabytes of Oracle storage or whatever thing. And it's planned. It's much more organized and therefore, it is much simpler and easier. And it also has the benefit for us that when a customer has a single PureArray, now when they get a second PureArray or a third or fifth or 10th. You just added into that service and it's far simpler and easier to bring new products on board. So that's the Enterprise Data Cloud.

Now what that does for a customer as well is it increases their ability to follow governance models, data sovereignty models, AI access restrictions or connecting up the data to AI both ways, making it easier to do that, but also easier to make sure that the AI is not operating on data. It increases their agility. And the biggest thing that's happening right now is the world is changing, right? AI is changing what people want to do with their data so rapidly. And the Enterprise Data Cloud gives them a ton of agility with that.

Now we're also going after AI with our FlashBlade EXA product, which is a new significant thrust. I would argue a lot of the hyperscale sales that we've had in interest is going more to AI than we had originally thought because again, the performance, the flexibility of the solution all helps with AI. So there's a whole lot we're doing down the AI path as well. But the Enterprise Data Cloud, the big thing it does increases agility, increases consistency, increases the ability of people to govern their data and their systems.

And I apologize it didn't go on me, but with the Enterprise Data Cloud that we're talking to here I think you had said, hey, we're starting with our own product, but over time, you'll actually allow other products to come inside that. How does that look? Or what is the timeline for that? I'm just curious.

The timeline for the other products is probably at least a couple of years out before you do much with that. We have a lot of great stuff to do with our own products first. And so we won't to do that. But when you think about the Enterprise Data Cloud and you think of the value that, that brings, yes, getting eventually your other resources in there is important to -- and we'll do things sooner like allow people to integrate with their backup products. And there are other things.

The other storage products will be a little further out. But being able to create a -- when you're creating a definition of a service level, you want to talk about your data protection and your backup and your cyber resiliency and your security and your location and such as well as the performance and other things you might think of as just being strictly with regard to a particular storage array.

And I know, I guess, we were talking about the AI opportunity. But curious, just given the number of neo clouds, if you will, that are starting to crop up where is Pure Storage as far as engaging and selling with these neo clouds? And can you also help us think through what that opportunity looks like? Because I imagine scoping and working alongside them is going to take a different angle versus a more traditional enterprise organization.

Yes. No, it is very different, and that's why we created the EXA product. The way we think about AI, you really have the hyperscalers who have all the layers of software. And for the hyperscalers, the product we sell in there, what we call our hydrogen appliance, which is really selling direct flash on an open compute or on their particular server in there is the approach that you want to take because they don't want to buy a packaged product. They don't want to buy something like a FlashArray or FlashBlade. They want to buy the closer to the raw bits, we'll call it.

And then on the enterprises, we have been strong. The FlashBlade S500 can support a good number of GPUs, it's the right kind of thing for them. The other thing I would say with most enterprises, people are not going to use AI by saying we're going to take -- every database we have every -- from health care, every patient record and everything and suck it out of the storage I'm currently using into some other AI training environment and go do everything there, right? They need the Oracles or the NetSuites or the SAPs or the Epic or all those things that are controlling that. And so they're going to in essence, used the FlashArrays and FlashBlades that they've been purchasing and just expect more performance and more flexibility out of them. And that's where we shine over some of the competitor areas.

So then the neo clouds and a few of what we call the tech titans who aren't hyperscalers, they fall in between that. And they want far more performance than a product like a straight flash blade is going to bring. And more scale. And that's where the FlashBlade EXA product is designed for.

And EXA, what we've done is we've said, okay, because this is moving so fast, you want to be able to take advantage of the hottest SSDs, the hottest performing CPUs, the hottest performing CXL, HBM, any of those things as they evolve. So FlashBlade EXA takes the metadata engine of FlashBlade and then has a whole set of data servers that are actually not our bots, they can be bring your own servers from the customer or a few server partners that we would specify with. And you can build up to almost any level of performance you want. The file system or the object store is controlled by the FlashBlade, the metadata lives there. And the data access goes directly to the data servers. And unless you scale performance of, like I said, practically any number you wish just based on how many servers you buy, that's the right answer for the GPU clouds. That's the right answer for the tech types that aren't quite hyperscalers.

And so I want to be clear here, too. I know that from a go-to-market standpoint, the management team has been targeting hyperscale type organizations. And so when we're talking about tech titans or neo clouds, is that falling within that hyperscale bucket? Is that what I should be thinking about. Is it a different...

A separate bucket.

That's okay. I want to make sure that we're -- Yes.

So think of the hyperscale bucket, there's there is like the five biggest hyperscalers who -- it's easy to figure out what those five names are. And then there's another 50, 100 sort of hyper -- second-tier hyperscalers we'll call it, that are far smaller, but they're not doing the GPU clouds we separate from those, from that second tier hyperscaler, right? So you have hyperscalers, the big ones and then the second tier, they have a different kind of approach that they want from the tech titans and the GPU clouds and then enterprises, right? And those are sort of the three segments we have.

So just to be clear, any time I say hyperscaler or anytime anyone Pure says hyperscaler, we do not mean the GPU clouds.

Okay. And just on that -- on the hyperscaler thing, you guys had announced a hyperscale win. This has been a big year as far as working through the co-engineering co-development. I think the latest update we received is that initially you expected to ship 1 to 2 exabytes. Now we are projecting more than 2 exabytes worth of shipments this year.

Can you walk us through how the relationship has progressed? What is driving that incremental volume? And how is the partnership evolving?

The relationship is going well. It's going as we expected. The deployment is going well. The -- I think like everyone else, AI is driving it even more the main hyperscaler and the Tier 2 hyperscalers we shipped to more stuff being driven by AI than we had originally thought would be, but everything is going well there. So we're happy with it.

Is that AI color, is that one of the reasons why the exabyte shipments are actually coming in above initial projections? Or is it, no, this partnership is actually growing faster than expected or sooner than expected. I'm trying to figure that out.

Yes. So I would say that -- you could say both.

Okay. And how should we think about -- I think you guys have been pretty consistent, but maybe for the audience, the scale or volume of shipments as we flip over into fiscal '27. I know there's been color commentary from the management team as we've gone through this process. But I do believe that there's a consensus view now that fiscal '27 we will get into production. So how is that shaping up? And just as a reminder, what should we expect in terms of volume commitments?

Given that I don't remember exactly what we've publicly said on that, and I'm not going to say anything about the volume shipments. And such on that. So listen to our earnings call for any updates.

Low double digits.

Okay. Sandra is reminding me, we said low double digits. I'll go that far.

Excellent. And then if I come to competition for a second, again, just given the evolution of the storage market and this need for where we are I think it's invited a host of new competitors, new entrants, and then there are others that are just meaningfully growing at scale. Are they -- are those new entrants actually coming up in competitive bake-offs more frequently? And is that in any way changing how Pure Storage is going to market? .

So I would say they're only coming up in very limited cases. So for example, the GPU clouds. There's a couple of -- there's like a competitor that's been around for far longer than Pure in there who's never been in the enterprise. And there's a couple of new ones and the GPU Cloud is a different environment. And in the traditional enterprise market, which is the bulk of our business, it's the same competitors. You hardly ever see one of the newer entrants there. And there, I think our competitors. Their biggest weapon against us has always been price or bundling their servers in with their storage and things. And as prices have gone up and they've all talked about increasing prices. I think that just it helps emphasize how much more efficient we are, right?

Our products have always tended to be smaller, denser, more efficient than their products. And so I think we're in better shape competitively at this point in time in that market.

Like I said, in the GPU clouds and sort of that aspect of the market, yes, that's where you see a bunch of competition from some of the new people. It's a little bit different because in that sense, we're the bigger, safer choice there as opposed to where the upstart in the enterprise space where the bigger one there, the GPU clouds. And so we'll get a share there, but others will get a share there, too. The products for the GPU clouds aren't really suitable for the enterprise.

And I guess when you do come against some of those less traditional competitors, let's say, do win rates materially change one way or the other? Or have they changed?

The win rates are good, but because it's sort of, let's call it, fewer deals right, overall, with fewer companies. You can't look at the win rates as being a stable, right? When I have 1,000 5,000 deals in traditional stuff, and I can sort of you can say, oh, our win rate is going to be x percent, yes, quarter-over-quarter, year-over-year, I can compare it. In the GPU clouds and things like that, it's just a far smaller number. And so therefore, one or two things slipping has a material impact on the win rates.

But I think our win rates are fine there. I think we'll do better because I like to be honest, FlashBlade X is a new product. And therefore, we're sort of in one sense, we've been serving this market a little bit less than others, and we're coming in and we're going to take a good share.

Great. And just on the product point as well. If I think about everything on and again, the demand for storage here, is it fair to assume that those evergreen on deals, are they actually increasing as far as size? Do you have visibility into that?

There's definitely a set of large ones, but there's also a bunch of smaller run rate stuff. I think when prices go up when availability becomes a question, I think it makes Evergreen/One look more attractive to people. I honestly think it's a better offering, and it should be more attractive all the time. But people look at it more when things are tight. When prices are high or when they can't get supply. And so yes, I think Evergreen/One will do well in the current environment.

Another element we haven't touched on yet is [ Kubernetes ] and Portworx to two separate pieces of the platform here. Is that resonating more with customers in the current environment, especially, I think about some industry M&A and the potential benefit or tailwind, if you will, for those two items in the Pure Storage platform.

Yes. No, definitely, it's resonating more. Everybody is looking at their alternatives for VMware and [ Kubernetes ] -- what we call modern virtualization [ Kubernetes ] and Portworx is -- I think, probably the best solution alternative solution for people. And it's a forward-looking one because it's a direction they already were planning to go towards containers. And so yes, that's definitely getting more traction.

And then probably the last question we'll have time for. I did get one from the audience, I want to make sure we're getting to. It's about the NVIDIA new rack design. And the question is if NVIDIA is bringing more storage to, I guess, VR deployment, what is the potential impact to Pure Storage?

Well, so the NVIDIA rack designs and things, again, they more go into the GPU clouds and things like that rather than the traditional enterprise storage. We're a storage partner of NVIDIA, just like some of the competitors that we see in the GPU clouds.

It's interesting. NVIDIA is evolving what they're doing so rapidly. It is difficult for everyone to keep up. So they come out with one of these rack designs before it can achieve real momentum, they come out with the next one and the next one because they're evolving so rapidly.

And so a number of these things that NVIDIA is pushing, I think you'll see effectively I'll call it fail in the market because they are moving on to the next thing so quickly. But we're a partner of there's, like I said, the competitors are a partner of theirs, and we're following along. And so I think as NVIDIA continues to come out with this. It's -- again, it's an opportunity for us to capture stuff in all of the GPU clouds and tech titans.

All right. Awesome. Well, we'll leave it there. But thank you to the audience, and thank you to the team here at Coz and Sandra. Thank you very much. .

Thank you.

Thank you, Mike.

Great. Good morning, everyone, and thank you again for joining the UBS Tech Conference. Final day. We're excited to have with us today Pure Storage. From Pure, we have Rob Lee, Chief Technology Officer. Before we get into the discussion I'm going to read the Pure Storage disclosures. Statements made in these discussions, which are not statements of historical fact are forward-looking statements based upon current expectations.

Actual results could differ materially from those projected due to a number of factors, including those referenced in Pure Storage's most recent SEC filings on Forms 10-Q, 10-K and 8-K. With that out of the way for Paul. Rob, thanks again for joining.

Thanks for having me, David.

So we just had earnings the other night. I think probably the best place to start is kind of what you're seeing in the business, how earnings, maybe just for people because everyone is a little bit busy this week with earnings in the conference. Maybe we can start there and touch on kind of the key takeaways, the highlights, how you're seeing demand? What was kind of notable in the quarter from your perspective?

Yes, absolutely. So yes, as you mentioned, we printed earnings 2 days ago, a really strong quarter. We beat on both top and the bottom line. And really, that's, I would say, a continuation of strength that we've seen throughout the year. If you recall, we similarly outperformed in Q2 and raised the outlook and guidance for the full year. Underlying that really, I would say, is broad-based strength that we're seeing across the enterprise business, across some of the newer product areas such as Portworx, we talked a bit about on the call. And then certainly, we had a discussion about starting to or really surpassing our previously anticipated full year shipments to the hyperscalers.

So a really strong quarter and really from what we see and flowing through to the outlook for the rest of the year, I continue to see that momentum building. I'd say that in the quarter and really the discussion that we had in the earnings are really 2 elements, I would say that -- well, maybe 3 elements that pervaded those discussions, right? One is commentary that we put out in terms of, hey, how to think about the incremental financial benefits accruing to us in terms of the hyperscaler business. Second was, hey, as we expand our discussions with our existing hyperscaler customer, other potential hyperscaler environments, really looking at, hey, what's the range and optionality in terms of additional business models that we can introduce.

And then third, as you would imagine, I'm sure you have questions on, a lot of discussion about commodities and kind of the supply chain tightness. And what I'd say is and maybe what a lot of the discussions we've had kind of post the call, what I'd say is on the incremental kind of spending commentary, really, what we're trying to say is, look, as the hyperscaler revenues come through, we see a ton of opportunity across the business, whether it's in the enterprise, whether it's going after newer markets, really going after the Neo-clouds, the AI market with FlashBlade/EXA with Vigor, certainly the hyperscalers. We see opportunities across the board to continue to invest for growth, and we intend to do that to a degree.

But we also are laser-focused on continuing to grow operating profit as well as to a degree, our operating margin expansion. And I think that, that's been a piece of the discussion that we've had post the earnings. And I would say the other thing that I would seek to clarify on the additional business models, right, is as you recall, our -- the existing business model with our first customer really has been framed around primarily a software license or royalty revenue associated with the shipments and deployment of the technology. We're not providing the customer any of the hardware components. Really, it's just a software royalty. As we discuss other environments of different price performance tiers with the customer, as we discuss engagements with other potential hyperscalers that we're in discussions with, what's become clear is, hey, there's a range of options, just like we tailor our technology solution to fit into their technology environment.

We have the optionality to tailor the business model and the mechanism for procurement into how it fits the hyperscaler best. But to be clear, we don't want to be in a position. It's not our intention to be in a position to be sourcing the full -- all the NAND for the hyperscalers. But we do see a range of options, premature to get into specifics. We see a range of options that really might, over time, take the gross margin profile associated with the hyperscaler revenue stream a bit closer to blended company averages, but we're not going to take the whole kind of NAND on the bottom.

Got it. No, that's a great overview and a lot to unpack there. So as you can imagine, in the last 1.5 days or so, we've been getting a lot of questions on that relationship that you just mentioned. So maybe we could start about what happened. Obviously, you outperformed on exabyte deployments. Can you help us understand how the customer, in this case, is thinking about that road map with you. I would imagine that deal that you currently announced previously is unchanged from a royalty perspective and a road map from an exabyte perspective. Is that a fair characterization of that current relationship? Or is there an opportunity to expand it to different models or different tiers of storage or use cases as you just described? Or is that more targeted towards other potential relationships with other hyperscaler environments?

Yes. So really 2 pieces in there, right? Hey, how is the relationship going? What is the current environment that we're serving? How does that grow? And then second, how does the new business model commentary layer into that? So let me hit those in 2 parts. So the relationship is going great. We -- as we've previously discussed, we're going to be a bit more circumspect about quantitative disclosure about shipments because of the wishes of the customer. But we had previously indicated coming into the year that we had expected and really contemplated in our guidance 1 to 2 exabytes of shipment for the year. We've surpassed that within Q3. And so we felt it important to update our financial community on that milestone check.

We've also said, hey, shipments continue. And we're -- our expectations that we've previously put out, we're not updating those. We're not changing those one way or the other. But the relationship goes great. And look, as we have discussed before, the way that hyperscalers in general design and deploy and ramp these types of environments, they go through a long design process to -- inclusive of storage, compute networking, HVAC systems, the whole line yards. They build a template. And once that template is defined for some period of time for a generation, if you will, as they have additional data center needs, as they have needs to decommission old data centers and refresh them, they'll go and deploy that template.

And so that's going well, and that's really the driver for scale. Now on the second part of the question, business models, as we are discussing -- as that first deployment is ramping well, and we've given the update on that, we are working with them to look at qualifying the technology into other price performance tiers. And as part of those discussions in the different environments, which, as you might imagine, involve different economics and discussions with other hyperscalers, that's really the catalyst for saying, "Hey, maybe there are other business models that might be more appropriate in those environments".

And was that a direct comment towards the existing customer that you have? Or was that relationship -- or that changing business model or the evolving business model originated from potentially a second or third...

Both.

For both, for multiple customers. Got it. Okay. And then when you think about the price performance dynamic that you just referenced, like what specifically -- I know you're not going to get into super nitty-gritty details until there's a deal announced, but when I look at the original deal with your first hyperscaler customer, it makes sense you bring to bear pretty significant technological IP with DFM, makes a ton of sense.

What are they looking at that you can bring to bear outside of that? I mean going on buying NAND like pretty basic skill set. So is it that the customer doesn't want to go in and actually manage those relationships with the memory providers, like what was sort of the genesis if you could speak to that or the origination of why the model might change at different price performance tiers?

Well, I think it just starts with the overall economics of the solution regardless of the piece parts, right? If you'd imagine at a higher performance tier, the technical makeup of the solution is going to change. We're going to use smaller drives, for example, configured for higher performance. They might fit into higher performance servers. So the economics of just the unit economics of that solution, that configuration are different. And when you look at the moving parts there, and there's a lot of moving pieces, not literally moving pieces around figuratively. When you kind of net all that out, that kind of opened the door to saying, "Hey, maybe there's another way to kind of structure this that makes sense for us, makes sense for the counterparty.

Okay. At the risk of you shutting me down. So I think Tarek said on the call, and you just reiterated. So when you think about your product gross margins going forward, it's more in line with how the business has been operated historically from -- don't extrapolate a royalty rate from a product gross margin perspective. Does that mean that sort of these new structural relationships are contemplated in sort of the preliminary comments that you made regarding '27 at this point? And if that's the case, then does that suggest that there's incremental either revenue from the existing customer or a second customer that could be in fiscal '27? -- from these types of relationships?

Yes. So look, we're premature to guide specific kind of puts and takes and incremental pieces to fiscal '27 revenue. But what I would say is, yes, the reason we brought the additional business models into the dialogue now is, yes, we are contemplating these different structures moving forward in a number of ways. We are contemplating potentially having different business models tailored to different hyperscalers and environments, right? But overall, right, as we deploy and introduce these new business models, we would expect the gross margin profile associated with the hyperscaler streams to approach a bit more closer to blended company averages from -- right now, they're very high, 90-plus percent. And that is kind of contemplated. And yes, I mean, if you just do the math, right, that would imply -- there's just naturally a trade-off in there between gross margin and top line.

And so we got this question a lot yesterday in just random conversations. But when you think about the competitive advantage that Pure has brought to the table, what in your view, from a technological perspective outside of the DFM capability kind of is extrapolated to future hyperscaler deals? Like what are they seeing that your -- that you can do that your competitors can't do beyond DFM at this point?

So I would -- what I'd say is the major differentiator and major pull is the DirectFlash technology. And that -- it does encompass the DirectFlash module, the physical module itself. But really, the lion's share of that IP is in the DirectFlash software that makes the module work, right? And my hardware guys will give me a hard time for saying this, but our secret sauce in that technology is not some magic in how we [solder] the chip into product...

It's an engineered product, okay.

It's the fact that we have the software to then go make that module highly reliable, capable of delivering 5, 10x the densities of hard disk drives and SSDs and performance at the same time, right? And the challenge is you can't -- with SSDs, you can't have all 3, right? You can build denser SSDs but you trade off reliability and performance, right? Like -- so that really, at the end of the day, is the driver of interest because, well, what does that get the hyperscaler, right? Yes, it's cool technology. But what it gets the hyperscaler is to say, "Hey, well, now I can serve the data needs that I foresee with the performance levels that they require in a much smaller footprint in -- with much less surrounding equipment, which then consumes less space, less power, which is in short supply and also has a greater increased reliability, right?

I just -- if I can ship a 5x denser solution, there's 5x less surrounding equipment that's 5x fewer things that can fail, right? And so that really, at the end of the day is what the hyperscalers are interested in. And then when they kind of tease it back and say, well, how do I get there? The directFlash technology is a critical enabler of that.

Got it. And maybe just one final question on this topic to start. When you think about the time line from conversation, project design, customer testing, I think in the past, you have said the original DFM win with the hyperscaler was measured in 12 to 18 months, correct me if I'm misremembering. Given the progress that you've made to date with this existing customer, does that allow you to compress the time line with future relationships, potential relationships? And could there be an opportunity to kind of obviously accelerate top line growth from these potential relationships because of all the engineering work, the design work that has been done and tested previously?

Yes and no, right? So yes, I think there are definitely phases of the process between knocking on the front door for the first time and booking revenue. There are phases of those processes that absolutely will accelerate. But then there are also gates and phases of the process that we have very little control over. And let me unpack that a little bit. A lot of the upfront work we did with the first customer was really convincing them the technology in TCO analysis and really in spreadsheet and PowerPoint form, if you will. Then you kind of go to, okay, great. this is promising. Let me go kick the tires, let me go do some light technology assessment. Okay, yes, it behaves the way I think it does. okay, now let's go understand how would I actually integrate this, a series of discussions, iterative design to go do that.

Oh, we have -- we think we have a way forward. Let me go proof of concept that. Let me go test that, et cetera, et cetera. And then that then joins into all of their other work streams to go figure out that design, right, not just storage, but again, compute, networking, rack design, power, HVAC. And then that whole thing has got to go then fit into their data center design cycle. And so to the degree that we've progressed with the first hyperscaler at scale now, they've been public to a degree about the solution. This is a small community. Everybody knows each other. That allows us to make progress very rapidly through some of those initial stages. And I think the integration work that we did with this first customer, what we have seen in subsequent conversations with other firms is, hey, the nature of this integration and the problem, the way that the hyperscalers are solving the problems at the layer above us are substantially very similar.

And so our belief is that 80%, 90% of that core IP can be reusable and transportable into the other hyperscaler environments. There's always going to be fit and finish, things like one firm is going to want to monitor systems a slightly different way than another and the way that they install and manage -- there's fit and finish work to be sure. But the core IP, we think, is very transportable.

Maybe just changing gears for a minute to the traditional core business. Obviously, you've been doing relatively well over the last several quarters. It feels like the storage market in key verticals, particularly in markets that you serve, maybe less so in the government, coming back a little bit stronger than maybe people thought. Can you talk to what you're seeing in that market ahead of -- we'll talk about commodity prices in a second. But what you're seeing in that market today from a demand perspective? I know we talked about it briefly the other day, but anything to call out from a vertical perspective, what customers are looking for solutions? What's driving sort of the traditional storage market demand strength that you're seeing today?

Yes. So I'm not sure I can call out specific verticals. Like we are seeing a pretty broad-based strength as we look back over the last couple of quarters. I would say in terms of customer segments, specifically in enterprise and kind of our most strategic pursuits. And part of that may be market and kind of nature and shape of demand. I think a large part of that is also our own focus, right? We've been very steadfastly focused on both technology portfolio, enterprise data cloud, the targeting of our sales force, like a whole number of things have really been focused on, hey, how do we go drive greater wallet penetration, wallet share penetration in large enterprise. And we're really starting to see those wheels turn.

I think you also do have some market forces that are driving specific demand in workload areas. We called out one on the call, which is modern virtualization. There's been some acquisition activity in the industry that have caused a lot of customers to want to look for virtualization options. And what we are seeing is the largest customers, big banks, Fortune 50 type customers are using this as an opportunity to say, "Hey, as I look for virtualization alternatives, instead of going to an alternate virtualization provider, maybe this is the time to go to open source, go to Kubernetes, use that to go manage my virtualization". And when they go down that road, they pretty quickly discover that Kubernetes plus virtualization pretty much equals Portworx.

And so we are seeing that, and that was -- that's one area where I think the shape and nature of demand has shifted a little bit. I think it's maybe early to call a trend, but I personally have been involved in more conversations this year than in previous years with the enterprise around selective repatriation, not broad-based, right? But some signs, right? Selective meaning, hey, there's this workload. We had put it in the cloud, but we, for a variety of reasons, now believe we want to bring that back for on-prem. But other than that, I wouldn't call out specific verticals. It is broad-based strength. And our focus has been and what we're seeing come through the business is, I would say, biased towards the enterprise.

And since you mentioned it, for maybe those less familiar, like what are you seeing from a customer perspective with regards to enterprise data cloud, right? Obviously, that's a differentiator for you. And that's been a relatively recent several quarters kind of dynamic. But what are you seeing from customers that is allowing you to win with that particular offering versus what the competitors are doing in the marketplace?

Well, so yes, let me touch a little bit on why we have gone down this road and why we see this as such a strong strategic pillar and then what we're seeing from customers. We've been, for many years, saying, "Hey, we provide tremendous advantages with our consolidated technology, one software approach, one hardware approach". That sameness, right, across the portfolio is one way to help customers break down silos. Historically, customers have a sprawling data footprint. They have point solutions for every need. And historically, each point solution may be from a different vendor, it's a different product. Each thing has to be managed separately. It's a lot of operational work.

We've had a tremendous advantage by having a sameness of our one software base, one hardware technology. So a customer as they buy more Pure has one thing to go and learn and manage. They don't have incremental work to go do. That's a critical driver of our as-a-service offerings. Well, if we take that one step further, right, and we say, all right, how do we then extend that operational simplicity to customers, that's what takes us down the Pure Fusion and Enterprise Data Cloud route. And customer feedback has been very, very strong. Customers want very performant, capable, reliable data storage. They don't want to spend a ton of time managing it, right? And that's what really Pure Fusion and Enterprise Data Cloud allows. Where we're headed with that is to not just automate data storage, but also the data management associated with that.

The other thing, and I touched on this a little bit on this conference call. The other thing we are seeing is it's allowing us now to have very meaningful conversations with different personas within the customer base. Historically, storage has been bought and sold by subject matter experts, the storage administrator, maybe a director, Senior Director, VP of IT, but it hasn't typically been a C-suite conversation, right? It's just -- it's kind of the plumbing, so to speak.

Well, now with Enterprise Data Cloud, what we can offer in terms of data set management, governance, security and really just the all-in view of where your data assets sit, it's actually driving us into much more meaningful and strategic conversations with the CIOs, the CTOs, the CISOs and again, taking us back to our focus on penetrating with greater wallet share, the largest enterprises, that really just helps us have a much more strategic approach as opposed to, oh, you've got this workload in the corner, let's go refresh that.

Got it. That's helpful. Obviously, top of mind, everyone here, you mentioned earlier, commodity supply chain. Obviously, the company has been through cycles. The company has extensive supply chain reach and expertise. And I think Charlie on the call touched on it or maybe Tarek touched on it, you know how to price. But maybe just as a refresher, in these types of cycles, this one may be a little bit different, how do you think about your price value relationship for customers? What do customers expect? Kind of what is your initial sort of reaction to the higher DRAM NAND prices? And ultimately, how do you think about the economic impact on, let's say, not just Pure, but ultimately, how do you think this plays out from an industry perspective, if you have a perspective?

Yes. No, absolutely. I mean, I would say, in some ways, the cycle may be different, but it's like every time I think it's different. There are a lot of similarities. Let me start by saying I've been with the firm over 12 years. I talked to a lot of customers. I have never once been in a customer conversation where a customer says, "Oh, commodity prices are doing this..."

I'm not buying.

"I want this on my system price." That connection just really isn't there. What we do see, right, in terms of pricing dynamics is, look, we price in a competitive market. We price at a premium to our competition. That premium reflects what we and really the market values our additional capabilities driven by software. And we typically price at a, call it, 15-point premium to the market. A lot of our competition are cost-plus players. And what we typically see is when we're competing head-to-head on our flash solutions and their SSD-based flash solutions, a rising commodity tide in terms of prices affects everybody, right? And what we typically see is that reduces my competitors' ability to aggressively price the solution, which then allows me to be more circumspect and disciplined on my discounting, right?

And so really, what we typically see is if commodity prices -- NAND commodity prices rise, the competitive prices will kind of -- the street prices will rise to pass that through. And we can be more disciplined in our discounting and a little bit more strategic in where we want to go win deals. The backdrop of that is because of our technology, because of our supply chain team, because of a number of things, we have structural and sustained advantages in all commodity markets, right? Because we use DirectFlash, because we don't source the same commodities they do, they're slightly different. We source NAND chips. They source transformed SSDs. They're slightly different ends of the same pool, so to speak. Those convey significant advantages. I mean, as an example, you take one of our smaller drives we ship 37.5 terabytes if you take that drive apart, you take the chips and you take a commercially available 30-terabyte SSD, take that apart, you'll see the same number of chips, the same chips potentially.

So why is it we can deliver 22%, 23% more usable storage out of that is because of the direct software. And so then when you take that into the commodity -- when you think about the cost of BOM, right, that's one example of how no matter what the commodity prices and markets are doing, we have structural advantages built in. And yes, so connecting this back to the commentary in the call, what we typically see is when we're competing flash to flash, we typically see the market prices adjust to meet those commodity prices. And therefore, we typically don't see a material impact on the product gross margins.

Right. And I think what we've thought in the past, and correct me if this is going to be different this cycle, customers allocate their storage budget to reflect the price. What I mean by that is they have a dollar amount or a fixed amount of storage that they're looking to buy for a particular workload or use case. Is that how you think this plays out where you're not going to see people spending, to your point, less dollars on storage. It may just look differently, maybe smaller capacity, a slightly different storage configuration, if you will. Is that fair?

Yes. I mean I think in the you can take things to the extremes, right? But I think in the short to medium term and within a reasonable regime, I think that is fair. I think they buy to their needs. It's relatively -- again, you take it to extremes and break down.

And then just maybe from a timing perspective, obviously, you're not a huge inventory build company. I know inventory was up a little bit, but not material. How should investors think about your ability to manage the supply chain from a duration perspective, given where NAND prices are in DRAM. So if we look out, let's say, I know you're not giving a guide for fiscal '27, but given sort of your purchase commitment dynamics, inventory, your relationship strategically, given where we are, let's say, from a spot price perspective, how do you feel about like the duration of your "protection" going into next year?

So what I'd say is we feel really good about where we sit, right? And I'll go back to what gives us confidence is a number of things, you outlined them, but also a very significant track record, right, of being able to navigate tricky supply chain situations and markets. I think, again, I'll go back to structural advantages we have. One of the other advantages we have, and this came out in the post-COVID tightness is because we share technology across the entire portfolio, we have ability to steer demand, right? And we have higher performance price value offerings. We have lower ones. So we have a lot of flexibility in terms of how we manage our inventory in our BOM. And we have a crack team kind of at the helm. We typically don't get on earnings calls and talk about strategic buys because we buy all the time and every buy we do is strategic.

All right. Just in the final minute or so that we have left, maybe I want to give the mic to you. Like what do you focus on? What are you strategically interested in, kind of what do you think the right takeaways are going into next year? I know there's a degree of maybe some confusion on what the market is going to look like, which we touched on from a commodity perspective. But what are the kind of takeaways that you think may be misunderstood about the Pure story going forward into next year?

Yes. Thank you for that. I think the thing that's probably least understood is how many facets of growth and opportunities we have in front of us. A lot of the dialogue from quarter-to-quarter is focused on hyperscalers or FICO focused on one area or another. And what's really not understood is, hey, we've got multiple areas of growth, and we're driving each of those in unison. When we look at the enterprise, right, we're something like a 13-ish kind of percent market share holder. We've got a long way to go there. Enterprise data cloud is a huge strategic thrust. Our focus on the enterprise is going to increase our penetration there.

When we look at hyperscalers, clearly, we've had a lot of dialogue there. We see that as an area of growth both within this customer, additional customers. We've talked about the Neo-clouds and our new product announcements and development to go penetrate that space deeper. We have industry trends that are driving growth such as modern virtualization. And I think what's not really appreciated is, well, how many vectors we're pursuing and really the opportunity set that sits behind it. At the same time, and I'll point the investors back to a discussion we had at the New York -- our New York Financial Analyst Meeting. At the same time, our ability to go pursue that is driven by significant IP leverage, right, which is to say we've assembled a very rich stack of intellectual property over the years in our enterprise offer that we're now able to go tease apart and repackage to meet the needs of multiple new market sets, whether it's the hyperscalers, whether it's the cloud native or whether it's AI.

And I had a deeper discussion of that and really the whole management team did in our New York materials, which are also on.

Great. Well, Rob, thank you very much for your time. It's been very generous. Thank you, everyone. And Andrew is in the back, if you want to hit for more questions.

Thanks.

Good day, and welcome to the Pure Storage Third Quarter Fiscal 2026 Financial Results Conference Call. Today's conference is being recorded. [Operator Instructions] At this time, I'd like to turn the call over to Paul Ziots, Vice President of Investor Relations. Please go ahead.

Thank you. Good afternoon, everyone, and welcome to Pure's Third Quarter Fiscal Year 2026 Earnings Conference Call.

On the call, we have Charlie Giancarlo, Chief Executive Officer; Tarek Robbiati, Chief Financial Officer; and Rob Lee, Chief Technology Officer. Following Charlie's and Tarek's prepared remarks, we will take questions.

Our press release was issued after close of market and is posted on our website where this call is being simultaneously webcast. The slides that accompany this webcast can be downloaded at investor.purestorage.com.

On this call today, we will make forward-looking statements, which are subject to various risks and uncertainties. These include statements regarding our financial outlook and operations, our strategy, technology and its advantages, our current and new product offerings, and competitive industry and economic trends.

Any forward-looking statements that we make are based on facts and assumptions as of today, and we undertake no obligation to update them.

Our actual results may differ materially from the results forecasted, and reported results should not be considered as an indication of future performance.

A discussion of some of the risks and uncertainties related to our business is contained in our filings with the SEC, and we refer you to these public filings.

During this call, all financial metrics and associated growth rates are non-GAAP measures other than revenue, remaining performance obligations, or RPO, and cash and investments. Reconciliations to the most directly comparable GAAP measures are provided in our earnings press release and slides.

This call is being broadcast live on the Pure Storage Investor Relations website and is being recorded for playback purposes. An archive of the webcast will be available on the IR website and is the property of Pure Storage.

Our fourth quarter fiscal 2026 quiet period begins at the close of business, Friday, January 16, 2026.

With that, I'll turn it over to Charlie.

Thank you, Paul. Good afternoon, everyone, and welcome to our Q3 FY '26 earnings call. Thank you for joining us.

Pure Storage delivered a strong Q3, continuing to expand revenue growth as customers increasingly look to Pure to solve their most pressing data management requirements.

Our results were underpinned by continued strength in enterprise and sustained momentum in our Evergreen//One and modern virtualization solutions, which includes CBS and Portworx.

During the quarter, we also exceeded our full annual forecast of 2 exabytes of hyperscale shipments and expect to ship more in Q4.

Our strong Q3 performance translates to an increased outlook for Q4 and improved guidance for FY '26, which Tarek will discuss in his prepared remarks.

Our enterprise momentum continues to be driven by the power of the Pure Storage platform, built on our Purity operating system, now enhanced with Fusion. Purity delivers the reliability, simplicity and long-term value that customers depend on to manage their data with confidence. It also powers our Evergreen subscription model, the industry's only continuously modern nondisruptive storage experience.

With Evergreen//One and Fusion, customers can build their own modern enterprise data cloud, automating storage, simplifying management and achieving faster, more efficient, lower-cost operations with zero downtime.

Since the beginning of the year, the number of customers deploying Fusion has more than tripled to the mid-hundreds, proof of the platform's momentum and market demand.

Data is now increasingly vital because of the promise of AI and requires that customers elevate its role in their technology architectures. While software may have been eating the world in the last decade, it appears that data will be eating the world and potentially even eat software in the next.

Since the beginning of modern computing, data has been structured below the applications that create it. It's been locked beneath databases, file systems and backup systems, each designed for a specific purpose, but ultimately isolating their data in silos confined to those services.

This application-centric model limits data visibility and mobility. It slows efficiency and innovation and prevents companies from realizing the full potential of their information. Consequently, data is repeatedly copied and transformed to be useful for other applications such as analytics and AI. Each copy is created and maintained by different individuals by manual processes.

With the massive proliferation of data and copies of data in enterprises managed by manual processes, data is poorly governed, often overproduced and highly fragmented. We believe the era of data being subservient to applications in data center architecture is ending.

Data, the lifeblood of modern organizations, must now take center stage in data center architecture. In a world where artificial intelligence, automation and analytics are redefining competitive advantage, enterprises can no longer afford to treat data as captive to specific applications.

Data must be architected to stand on its own, self-describing, stateless, and managed globally by policy set in software. The Enterprise Data Cloud makes this possible. It gives organizations the ability to access and leverage all their data securely, seamlessly and in real time, regardless of where it originates.

With the right authorizations, any application will be able to access integrated pools of data, enabling faster insight, more intelligent decision-making and greater operational velocity in business.

By freeing data from legacy silos, the Enterprise Data Cloud lets companies operate with the same flexibility, scalability and efficiency as the cloud itself.

Customers who use our Pure Fusion capability embedded in Purity can now manage their data sets globally with policies embedded in software, rather than by fingers on keyboards, enabling storage and data management that is truly defined by software.

At our Accelerate roadshow in New York and around the world, we extended the Enterprise Data Cloud into Azure with our Pure Storage Cloud, which enables customers to unify their data landscape across public and private environments. This Azure-native service for AVS enables seamless migration from on-prem VMware environments with enterprise-grade resiliency and efficiency.

In New York, we continued rolling out powerful innovations across all 3 pillars of the Enterprise Data Cloud. First, we expanded our unified data platform with new systems like the XL 190, and we improved data reduction efficiency on all our platforms, giving customers more capacity and performance on their existing systems.

Second, we enhanced our intelligent control plane with an AI Copilot, which simplifies management and automates complex tasks, making storage operations faster, smarter and more reliable.

Third, we expanded our partner ecosystem to deliver greater value through integrated cybersecurity and data protection. As the pace of our technology advancement accelerates with scale, we expect to continue to gain market share in more and more segments of the data storage and management space. In the quarter, we were recognized in the two most important Gartner Magic Quadrants for our industry. In the Enterprise Storage Platforms Magic Quadrant, Pure was positioned highest for execution and furthest for vision. We were also recognized as a leader in the first-ever Infrastructure Platform Consumption Services Magic Quadrant.

Additionally, Pure was recognized as a leader in the IDC MarketScape on support services globally, reflecting our strengths in reliability, proactive connective support and our customer-first mindset.

Portworx continues to lead the industry in defining storage in the cloud-native, Kubernetes and container world. Customers want more flexibility, lower cost and modern architectures that support cloud flexibility. It's why companies like NVIDIA, SiriusXM and a major global bank have chosen Portworx. In Q3, one of the world's largest enterprise software companies selected Portworx to overcome multi-cloud fragmentation. They accelerated the deployment of their cloud services across AWS, Azure, Google Cloud and Alibaba Cloud, ensuring a consistent operational experience, enterprise-grade data protection and high availability.

Modern virtualization is a subject in great demand with our customers. Across the industry, 3 trends are driving this shift. The search for alternatives to expensive legacy virtualization models, the rise of containers and KubeVirt, and the significant increase of AI and machine learning built on Kubernetes.

Portworx and our solutions in partnership with Nutanix, Microsoft, Red Hat and others are leading this transition away from traditional virtualization solutions. Portworx is now becoming practically mandatory for any scaled Kubernetes virtualization deployment. As Kubernetes extends beyond virtualization to power modern applications and AI workloads, Portworx's role continues to grow. Portworx lets customers run any application anywhere, securely, efficiently and up to 2 or 3x lower cost, so they can modernize faster and operate with greater speed and flexibility.

Neoclouds represent another fast-growing market for specialized storage technology. This new generation of specialized high-performance cloud platforms built for AI, machine learning and other compute-intensive workloads represent a new segment of cloud infrastructure, driving new benchmarks for performance and scale. Recently, we published our latest benchmarks for FlashBlade at Supercompute, a leading conference on high-performance computing and AI.

FlashBlade//EXA delivered data to thousands of GPUs twice as fast as competing systems in less than half a rack. FlashBlade//EXA extends the power of our Purity architecture to these next-generation clouds, pushing the limits of performance for AI and high-performance computing with superior sustainable throughput and scalability.

As we have discussed over the past year, Pure provides a compelling alternative to hyperscalers who face mounting hard disk and SSD cost and power constraints. As I stated earlier, we have already exceeded our annual plan for shipments by the end of Q3. But consistent with our statements at our September Financial Analyst Conference, we will not be providing specific information on shipments to hyperscale customers going forward. We will share more information next quarter about the outlook for FY '27 and the economics of our hyperscale business as it impacts our financials.

Turning to the macro environment, we foresee increased commodity pricing and excess demand putting pressure on global supply chains. As in the supply chain crisis of 2021 and 2022, we anticipate both extended component lead times and higher component pricing across the technology industry in the quarters ahead. Pure is well prepared for this challenge with a resilient supply chain, a broad global supplier base, manufacturing sites on 3 continents and strong business continuity plans.

As we've noted a number of times before, given our industry's dynamic pricing environment, the effect of commodity pricing tends to affect our top line more than gross margin. Thus, we would expect higher commodity pricing to positively affect revenue growth.

Finally, I am pleased to welcome Pat Finn to Pure as our next Chief Revenue Officer. Pat brings extensive experience in scaling sales and go-to-market organizations within high-tech infrastructure companies, along with a proven record of building lasting customer relationships with leading global enterprises.

I also want to extend my gratitude to Dan FitzSimons for his dedication and contributions to Pure over the last decade and for his continuing engagement with Pure to maximize our opportunity. His leadership was and is instrumental in expanding our operations, advancing our enterprise and commercial strategies and helping advance Pure from its early days to the global enterprise it is today.

With that, I will hand it over to Tarek.

Thank you, Charlie. In Q3, we delivered strong revenue and operating profit results, both exceeding the high end of our guidance range.

Revenue of $964 million grew 16% year-over-year and operating profit grew 17% year-over-year to $196 million which is a record for the company and resulted in an operating margin of 20.3%.

Our Q3 results demonstrate sustained demand for our differentiated data storage and management offerings. Sales across our portfolio remained robust, led by ongoing strength in the enterprise and our hyperscaler business. We also continue to see strong traction of our Evergreen//One and modern virtualization solutions, which includes Cloud Block Store.

Our success this quarter continued to be driven by the strength of the Pure platform value proposition across multiple customer segments. Our single operating system, Purity delivers simplicity and reliability across just 2 hardware offerings, blades and arrays. We provide the industry's only truly nondisruptive upgrades and the only genuine storage as a service offering through Evergreen//One.

In terms of data management, no competitor can match the capabilities we are delivering with Fusion and the Enterprise Data Cloud. And with our differentiated DirectFlash technology, we are able to expand into a large, newly addressable hyperscale market, where traditional storage system vendors cannot compete.

Underscoring this competitive advantage, hyperscaler shipments as of Q3 year-to-date exceeded our original forecast for fiscal year '26 of 1 to 2 exabytes. We expect momentum in our hyperscaler business to continue in Q4 and obviously for fiscal year '27. We will not be providing any additional quantitative guidance on our hyperscaler business this fiscal year, but we do expect to provide additional color for our next fiscal year at the end of Q4.

Product revenue of $534 million grew 18% year-over-year. As a reminder, our product revenue category now includes royalties that we receive from hyperscale shipments as well as a portion of Portworx software revenue when sold as term licenses. Q3 TCV sales for our storage-as-a-service offerings grew 25% year-over-year to $120 million. This consistent growth reflects Evergreen//One and subscription-based offerings, strong resonance with our customers by delivering a consistent, nondisruptive operating and management environment.

Subscription services revenue in Q3 reached $430 million, up 14% year-over-year, accounting for 45% of total revenue. ARR grew 17% to $1.8 billion, while total remaining performance obligations, or RPO, grew 24% to $2.9 billion. RPO encompassing our storage-as-a-service offerings and Evergreen subscriptions across our installed base grew 22% exiting Q3.

With respect to our geographic mix of revenues, U.S. revenue was $683 million, growing 22%, and international revenue was $281 million, growing 4% year-over-year. Overall, we added 258 new customers and our penetration of the Fortune 500 is now 63%.

Turning to margins and profitability. Total gross margin increased to 74.1%. Subscription services gross margin was 75.5%, and product gross margin increased to 72.9%.

Growth in product gross margins reflects a stronger mix of higher performance flash arrays, a slightly larger proportion of Portworx software sold as term licenses as well as hyperscaler shipments. On a full year basis, we expect that product gross margins will sit closer to 70% with some variability in magnitude quarter-to-quarter.

Operating profit of $196 million and operating margins of 20.3% in Q3 were both positively impacted by revenue strength and robust gross margins.

Our head count increased sequentially by 104 employees to approximately 6,200 employees.

Our balance sheet remains strong with $1.5 billion in cash and investments. Q3 operating cash flow was $116 million and our capital investments of $63 million included test and infrastructure equipment to support data center expansion and funding of Evergreen//One subscription growth.

In Q3, our free cash flow performance was strong as we generated $53 million of free cash flow for free cash flow margin on revenue of 5.5%.

We returned $53 million to shareholders through the repurchase of 600,000 shares and offset roughly 1 million shares in employee's award withholding taxes, and we currently have $56 million of our buyback authorization remaining. We intend to update you on a new share repurchase authorization at the end of our fiscal year '26.

Now turning to our guidance for fiscal year '26. Strong Q3 results and higher expectations for Q4 contributed to an increase in full year revenue and operating profit guidance for fiscal year '26.

For Q4, we anticipate revenue to be in the range of $1.02 billion to $1.04 billion, representing approximately 17.1% year-over-year increase at the midpoint. We also expect operating profit to be in the range of $220 million to $230 million, representing approximately a 47% year-over-year increase at the midpoint.

As a result of our Q4 guidance, for fiscal year '26, we anticipate revenue to be in the range of $3.63 billion to $3.64 billion, representing 14.7% year-over-year growth at the midpoint. This is a 70 basis points increase from our previously provided revenue guidance of 14% year-over-year growth.

We expect operating profit to be in the range of $629 million to $639 million, representing approximately a 13.3% year-over-year increase at the midpoint. This is over a 330 basis points increase from our previously provided operating profit guidance.

The projected increase in operating income for fiscal year '26 reflects the strength of our business and the impact of hyperscaler revenues on gross and operating margins. Most importantly, I would like to emphasize and reiterate that beyond fiscal year '26, we are planning to capitalize on the financial benefits from hyperscaler revenues to continue making significant incremental investments in R&D and sales and marketing in order to sustain our momentum and capture additional profitable growth opportunities in the enterprise aligned with our long-term strategy.

As we close fiscal year '26, we will provide guidance for fiscal year '27 that will factor in these increased investments.

In addition, and as foreshadowed by Charlie, we plan to grow our hyperscaler business. In doing so, we will be evaluating additional business model options that may result in changes in gross margin economics for the hyperscaler business in fiscal year '27 relative to fiscal year '26. We will also provide an update to the market as we finish fiscal year '26 and provide guidance for fiscal year '27.

With that, I'll now turn the call back to Paul for Q&A.

Thanks, Tarek. [Operator Instructions] Operator, let's get started.

[Operator Instructions] Our first question comes from Amit Daryanani from Evercore ISI.

I guess there's been a lot of focus on memory price inflation across both NAND and DRAM and certainly your product gross margins at near 73% don't suggest this is an issue in October.

But can you just talk about how should we think broadly both on the direct impact on your P&L from a margin and revenue perspective from this commodity inflation? And then also maybe just touch on the indirect impact you could see from competitors' pricing changes as you go forward.

You bet. Thanks, Amit. Well, I think many of the analysts who have followed us for many years have come to understand is that in our market, which has a very dynamic pricing environment, prices are set at the time of purchase, and we compete with other vendors, many of whom are cost-plus vendors. And as such, the pricing of our systems tends to float more with commodity pricing. What that means is that it tends -- the commodity pricing tends to be a lesser effect on our gross margins and frankly, a greater effect on the overall market, that is, when prices are higher, to the extent that customers require the same amount of capacity, they'll be paying higher prices on average.

So this is something that customers understand. Sometimes customers will buy ahead. But what we expect is that higher prices would generally translate to a rising tide, if you will, in the storage market overall, of which we'd be a beneficiary.

So yes, 73% is a bit high, as you're pointing out. Generally, our long-term trend, we believe, is in the 65% to 70% range. And that's the way we'll continue to guide for the foreseeable future.

Our next question comes from Aaron Rakers from Wells Fargo.

I guess maybe building on Amit's question a little bit. I think that the one number that stands out to me in this print is like you had a 76% sequential increase in inventory.

I'm curious if that is reflective of any kind of strategic purchases, appreciating any color you can give us on how much maybe purchase obligation stands coming out of this quarter or does that not necessarily relate to the component pricing environment? Is it more of a reflection of your procurement for your hyperscale customer? Any color there would be helpful.

Well, let me start and then I'll pass it over to Tarek. We -- as you may know, Aaron, we don't crow about making -- about our purchase decisions being ahead or on the spot market on a quarter-by-quarter basis.

We do believe we've got one of the best supply chain and purchasing organizations, frankly, in our industry. But we're always taking advantage of what -- of our own internal forecast of what the market will be like.

I'll have Tarek comment on the specific numbers.

Yes, Aaron. Thank you for the question. Look, we track and manage inventory levels closely, and there are a couple of drivers in the increase you referred to. First, we had tariff mitigation purchases at the beginning of the year and also the hardware components market is tightening all over the world and we've taken a couple of positions in some key parts to avoid disruption in our supply chain, which is why you could see that increase.

But overall, I would say that our inventory levels at about $46 million in Q3 are low compared to the overall size of our business. And we think that it could stay about the same level moving forward.

Our next question comes from Howard Ma from Guggenheim Securities.

If we make an estimate for hyperscaler shipments in the quarter, the implied product gross margin, I believe, was still quite strong and perhaps a notable sequential uptick. So my question is, was there a continued mix shift to higher-end products in the quarter? And what is your ability to maintain this level of product gross margin in coming quarters, especially when considering potential sales to neocloud, sovereign cloud that I believe probably will carry lower gross margin.

Yes, Howard, thank you for the question. It's Tarek here. There are 3 vectors that drive the overall product revenue gross margin.

Number one, the product mix in itself. And we did see higher and greater configs being bought by customers in Q3, and this is usually good news for product revenue gross margins.

The second element that drives the product gross margin is, to a large degree, purchases of Portworx licenses that are purchased on a terms basis. Sometimes customers choose term licenses instead of subscriptions, and we did recognize some Portworx revenue in product revenue this quarter in Q3.

Thirdly, obviously, there is a contribution from hyperscalers revenue, which we recognized in Q3. As we mentioned earlier on, we exceeded the prior goal of 1 to 2 exabyte shipments for fiscal year '26, Q3 year-to-date, we already attained that goal, and we can see some additional growth in Q4 and beyond that point.

Our next question comes from Mike Cikos from Needham & Company.

This is Matt Calitri on for Mike Cikos over at Needham. I'm wondering how early feedback has been on the Enterprise Data Cloud with Pure Storage Cloud and how this expanded offering is impacting the way that customers are approaching their storage architectures?

Yes. It's still early days. We're in the first year. And -- but the response has been extraordinarily positive. We are seeing a lot of demand for placement or for the feature to be placed on the systems that already exist. That they have to go through a software upgrade to do that. And since the release of what we call our long-lived or enterprise release, which, of course, some of the larger companies will wait for -- since that time, it's been expanding very rapidly. So -- and now that we've released the Pure Storage Cloud, which is effectively, if you remember, our Cloud Block Store. It's effectively Cloud Block Store but completely cloud native in Azure and managed by us. So it is, in fact, a cloud service. And it's part of the Enterprise Data Cloud.

So now customers can very easily manage their global estate, including what's in Azure now on Pure Storage Cloud under the EDC umbrella. So this is -- we believe this is part of what's driving customers in our direction. You saw our growth this quarter, well above the industry as a whole. And we think that the vision and the capabilities that we're bringing with Enterprise Data Cloud is a significant portion of that.

Yes. And Matt, this is Rob. Just to add on to what Charlie said. I think one of the things that's been quite notable in terms of early customer feedback with Fusion -- Pure Fusion and Enterprise Data Cloud is both not only how positively it's being received, but how it's being received by multiple persona sets within our customer base.

As we've discussed in prior calls, we have historically interacted with subject matter experts, storage administrators and the like within our customer estates. That group of customers really is gravitating towards storage automation, the ease of performing operational tasks that used to be quite manual in nature.

But as we further articulate the Enterprise Data Cloud vision, we're now having very meaningful conversations with Chief Information Security Officer, CIOs, folks that are really recognizing the benefits of governance data security, data provenance that we can go and deliver.

So as Charlie has mentioned before, one of the things that Enterprise Data Cloud is allowing us to do is now have much more strategic interactions with our largest and most valuable customers.

Our next question comes from Samik Chatterjee from JPMorgan Chase.

I know you're not guiding quantitatively to the hyperscaler sort of outlook here for next year. But wanted to see if you can give us a sort of update on what the engagements with the hyperscalers beyond the first customer that you have are progressing?

And when you talk about further investments to capitalize on those revenue opportunities, if you could flesh that out a bit in terms of are these investments in sort of more product SKUs? Or what are the nature of these investments that you're looking to ramp into next year?

Yes. Samik, this is Rob. I'll take that one. Look, as we continue to progress with our existing hyperscaler customer and work with additional ones, those activities are progressing well. We continue to be engaged with the majority of, let's say, the top 10 hyperscalers in one way or another with multiple proofs of concept that are being undertaken and are underway.

As we look at -- I think the second part of your question, correct me if I'm wrong, was areas of investment that we'll be looking at across the board. So certainly, we have continued investments that we've discussed around our DirectFlash road map, driving greater densities that continue to allow us to penetrate wider swath of price performance tiers.

But then as we look at the enterprise portfolio as well, when we look at delivering on the Enterprise Data Cloud vision, if we look at AI, what we're doing with FlashBlade//EXA, there's a number of areas where we'll be directing that continued investment to drive growth in not just the hyperscaler segment of the market, our enterprise market, but also AI and neoclouds.

Our next question comes from James Fish from Piper Sandler.

Could the increase in memory costs actually lead to a change on the demand dynamics between essentially your OpEx and your CapEx model in your view?

And can you just remind us or walk us through for every 1% or 10% increase in commodity costs, especially on the flash side, what that leads to in terms of that revenue versus gross margin delta that you guys are saying, like, hey, it doesn't impact us much. I get why, just trying to understand the magnitude here.

Yes. So let me take a stab at part of the question. Generally, we do -- when economics in the market change, whether that's interest rates or changes in expectations on the pricing side, you're going to see shifts between Evergreen//One and the -- and our -- the traditional product purchase. So I would expect it -- now we've been fooled before, to be honest, but I would expect it to -- that are increasing prices to drive more -- relatively more customers towards the Evergreen//One model.

The other part, though, I would say is that because we have longer depreciation cycles on the Evergreen//One model, it really insulates us quite a bit from spot pricing in the commodity market.

So I think, again, that's another place where we're also able, as you may recall, to use existing and returned equipment that has been renovated in the Evergreen//One model. So there's always a mix of new and existing and renovated product that goes into the Evergreen//One environment. Again, that isolates us, if you will, from spot pricing.

Our next question comes from Simon Leopold from Raymond James.

This is Victor Chiu in for Simon Leopold. You noted in your prepared remarks that scaling the hyperscale business could change the gross margin dynamics. Can you expand on this a bit?

I'm presuming that you're expecting a structural expansion shift in gross margin as you scale that business out. Can you help us understand kind of the leverage around that and what your expectations would be based on what you're expecting for the next year?

All right. So first of all, I would say you have to look at product gross margins, excluding hyperscalers. We guided in the past that our typical product gross margin range is between 65% and 70%. If you think about the nature of what we sell to hyperscalers, right now, we recognize only software licensing revenue, which is very high gross margins. As we said on prior occasions, this is a 90-plus percent gross margin revenue stream. And therefore, obviously, as you combine the two, traditional product revenue and software licensing revenue, this has a very positive impact on gross margins moving forward.

Having said that, I also said in my prepared remarks that we are exploring new and different revenue models with hyperscalers. In addition to recognizing the net revenue, we may think -- we're thinking right now that in the future, we will be recognizing different type of revenue streams than licenses, and this would have an impact on the gross margin economics of the revenue stream moving forward in fiscal year '27. I will update you about this in our Q4 call. So for the moment, no change.

Our next question comes from Wamsi Mohan from Bank of America.

I was wondering, when you look at sort of your overall results, you obviously exceeded your forecast of 1 to 2 exabytes at hyperscalers, which should be coming in at very high margins. But you have the lowest magnitude of earnings beat in a while.

And so as we think about that, probably it's coming from the higher investments that Tarek alluded to. And so as we think about this sustaining into fiscal '27, does it really mean that we're going to see a year where potentially operating margins could be down year-on-year. Just wanted to understand, Tarek, you really seem to emphasize that point and just want to make sure that we walk away with the right conclusion around trajectory of margin.

Well, Wamsi, let me start, and then I'll hand it over to Tarek. You may recall that as we entered this year, we actually guided to a flat operating margin business for the year in anticipation of higher investments in the hyperscale business that was without a full understanding of what the revenue would look like in that from the hyperscale business.

As we are here sitting today, it is a substantial increase over our original guide. So we're coming -- we'll be coming in -- our expectation is higher than our original guide, which is a positive -- which is very positive. And we stay committed to continuing to increase operating margin on a year-over-year basis. So that shouldn't be under -- there shouldn't be any question of that at all. Tarek, do you want to...

Yes. Thank you, Charlie. Wamsi, I'd say the following to you. First of all, with respect to fiscal year '26, the new guide implies a pretty substantial growth in operating profit. I mean if you really look at it on a year-over-year basis in Q4, at the midpoint, the operating profit growth implied is 47%.

When I joined Pure almost 6 months ago, we observed that there were several opportunities in the enterprise space that we wanted to capture. And we can see substantial growth there, and we intend to continue to deliver the levels of growth that you're seeing in Q3, in Q4 with the guidance that we implied, but also extending that into fiscal year '27.

But to go after that growth, we have to make some investments in sales and marketing, in R&D and also in back-office systems to accelerate the velocities of our deal constructs and delivery. And it is for this reason that during the Q2 earnings call, we changed our guidance philosophy to guide on operating profit growth with a range as opposed to a single operating profit margin figure. However, as Charlie mentioned, we continue to expect operating profit to grow beyond fiscal year '26. And yes, you can expect a degree of operating profit margin expansion in fiscal years beyond fiscal year '26.

Our next question comes from Erik Woodring from Morgan Stanley.

I'm going to go back to the kind of memory questions that a lot of us are asking. And Charlie, I would just love to get opinion on why or if demand elasticity in your markets could be different this time versus 2017 or 2021?

I'm just trying to think of looking back at memory cycles, it's clear that you were able to kind of pass this pricing through. Could that be an inhibitor to market demand this time around? And if so, why or why not?

Yes. It's actually a great question, Erik, because obviously, the 2 factors go together, right, which is to say that just because prices go up doesn't mean that customers have an unlimited ability to pay. And so demand may go down.

What we've seen overall though, in the years -- and you're right, 2017, roughly 2021 was that commodity pricing tended to have a much greater effect on the top line or you might think of it as the overall market size or growth in the flash storage industry than it did at least on our gross margins overall.

We're able to manage the gross margins actually quite well. And part of that is because we do compete with competitors that largely sell on a cost-plus basis. And that tends to then float pricing along with commodity prices.

There's not much more than that. You're right, it could constrain demand a little bit from an overall terabyte standpoint, but the dollar -- generally, the dollars continue to scale along with commodity pricing. I think the main thing, of course, is with AI, we are seeing more demand for data just in general. And data has become and data architectures have become more top of mind for our customers. So I don't see -- I think that will continue, even though, as you point out, budgets are not necessarily unlimited.

Our next question comes from Krish Sankar from TD Cowen.

Tarek, I understand you're going to give us more color on the next earnings call regarding the hyperscaler business model shift. I'm just curious from the economics of this opportunity standpoint, is the rising land prices one of the catalysts for the shift where the hyperscalers don't want to deal with the volatility? Or in other words, if 2 years from now, if NAND prices do crash, will the business model again shift?

I would say -- let me jump in. I would say that at the moment, that doesn't seem to be a big factor in this. I would say that we're just seeing different hyperscalers have different points of view as to how they'd like to purchase. And given that there's going to be a different mix of options that carry different COGS associated with it. And that's going to be a large part of us having a somewhat more mixed and nuanced model in the hyperscale business. It is certainly true that the hyperscalers in some ways, would rather be more insulated from commodity prices, but there's only so much that -- we could do some of that, but we can only do so much of that.

Our next question comes from Asiya Merchant from Citi.

Charlie at the Accelerate event in New York, I think I asked about the neocloud opportunity. And I think the response there was, it's a little bit more nuanced and relative to the size of the hyperscaler opportunity, of course.

Can you just double click a little bit about how you're thinking? Has that thinking changed about how you're looking at the neocloud opportunity ahead? And does it relate to any of your product offerings lately?

Thank you, Asiya. We try to be very structured in terms of how we qualify hyperscale opportunity and how we qualify, for example, neocloud and AI opportunity. And let me just repeat that and then I'll go directly to your answer.

So when we sell into the hyperscale into their traditional structured storage environment, much of which is used for AI, but it's not specialized for AI, we discuss that as being the hyperscale opportunity.

When we sell specialized product that's designed specifically for AI, which comes under our FlashBlade product traditionally and typically the FlashBlade//S, but now the FlashBlade//EXA. And that is what's typically sold into the neoclouds, sovereign clouds and other AI environments. We discussed that as being the AI -- our AI market.

So just to put those two in -- make that clear because a lot of the products we will be selling into a hyperscale, despite the fact that it's going into their standard storage environment, will be used for their AI systems. In fact, they tend not to buy a specialized product for -- in the storage space for AI. So in that sense, it's not nuanced.

Obviously, we're a little bit further along, if you will, on having really scaled sales into the AI environment. We've been selling FlashBlade//S now for 5 years. We have hundreds of customers using it directly tied to GPUs for AI. With FlashBlade//EXA now, they can scale to even larger sizes. And we're seeing a good interest right now in FlashBlade//EXA in the neocloud environment. So does that answer your question, Asiya?

We'll take that as a yes.

Our next question comes from Eric Martinuzzi from Lake Street Capital.

Yes. I wanted to -- based on the investment comments that you had, Tarek, around the R&D and sales and marketing, just -- for those of us modeling at home, that mix, the skewing of the investment, just curious to know if there's an expectation that next year would be significantly different than this year.

In other words, based on my math, I've got about 34% of your non-GAAP operating profits in the first half of FY '26 and 66% in the back half of FY '26. Is the expectation that, that would skew abnormally in FY '27, such that it would be more back-half loaded?

Look, it's -- we're still in the middle of our planning cycle and we have yet to give guidance for fiscal year '27, and we'll do so in Q4.

But I think the point to take away here is that we would like to continue to grow to the levels that we're witnessing right now in fiscal year '26 into fiscal year '27 and beyond. And therefore, this growth comes at a cost, and this is in R&D, in sales and marketing, and in back-office system operations.

But if you are looking at the various OpEx lines as a percentage of revenues, there won't be a material disruption to the levels that you're seeing today in R&D, sales and marketing as a percentage of revenues. But the dollars overall will increase.

We have one -- this will be the last question coming up next.

Our last question comes from Mehdi Hosseini from Susquehanna.

Yes. Charlie, I want to go back to that -- to exabyte shipments. When I look at the overall size of the enterprise SSD market, we're tracking close to about 400 exabyte.

And my question to you is what would need to happen for you to scale and actually account for 1% or 2% of the exabyte shipment, especially given the fact that hyperscalers are driving the growth for enterprise SSD demand. And let me know if my question is not clear.

Well, I think I understand what you're asking, which is what would it take for us to continue to grow in the hyperscale business to become a much more substantial portion of their overall storage purchases, which today is hard disk and SSD.

Our view is that we provide a superior solution in both environments, that is hard disk and SSD. And now, of course, the design cycle within the hyperscalers is quite long. You have to, first of all, meet the beginning of their design cycle and then their design cycle is about 2 years.

And so -- and we are a different -- to be clear, we don't operate in the same way as an SSD or a hard disk. And therefore, it requires a change in their technology and operating process.

That being said, as Rob mentioned earlier, we're in a lot of POCs. We're in a lot of engineering meetings with the majority of the top 10 hyperscalers. The value that we provide in the area of power and space reduction, in the area of performance increase and in consistency. Consistency because whether it is the lower performance, low price range or the high performance and therefore, higher price range, we provide them a single solution from a software perspective that doesn't require them to modify their operating systems.

These are all very powerful incentives for them to make that investment and make that change. So our belief is that over time, obviously, we're going through right now a supply chain crunch that is that the hyperscalers are seeing that supply chain crunch. And every form of memory and every form of storage now is basically on back order.

And when we see that, it does give us an opportunity to get in, but we think that when that -- when supply chains come back to normal, which always occurs, there's always a reversion to the mean, it will have us be in place with superior economics and performance. So we do believe that we can continue to take share and become a significant portion of hyperscaler storage environments.

Yes, Mehdi, I would just net it out to 3 things. One is making our existing customers successful in the environment and tier that we've been designed into. Two is expanding our offering and value to multiple tiers of price performance. And three is expanding the reach of our technology into other firms, right? And so Charlie talked about a lot of the process and steps along that path. We're pursuing it afoot. And -- but that's how I think about what are the unlocks and what are the drivers to get there?

Charlie has some concluding remarks.

Yes, I want to thank you and thank everyone for joining us today. Data is now becoming, we believe, the primary engine that's going to drive economic growth in the decades ahead. And it needs to be elevated to be useful and available for a wide range of uses.

And the Enterprise Data Cloud makes that possible. It gives customers a unified and virtualized way to control their global data estate with the reliability and the simplicity that we, Pure, are known for.

So once again, as always, I wish to thank our customers, partners, our suppliers, our employees and our investors. Your consistent support drives our success. Thank you.

That concludes the Pure Storage Third Quarter Fiscal 2026 Financial Results Conference Call. Thank you for your participation. You may now disconnect your lines.

Welcome to the Pure Storage Product & Technology-Focused Meeting for Financial Analysts. Let me remind you that we will be making forward-looking statements today that are subject to assumptions, risks and uncertainties. Actual results could differ materially from those anticipated due to a number of factors, including those referenced in the detailed disclaimer at the beginning of our presentation slide deck and in our public filings with the SEC, which we encourage you to review.

The presentation slides discussed today will be available on our Investor Relations website at investor.purestorage.com.

Hello, hello, and welcome to our product and technology-focused Financial Analyst Meeting, which is being held in conjunction with Accelerate. We're so happy to see so many people here in person as well as those attending virtually. We really appreciate your interest in Pure Storage and your support, frankly, many of you that are here.

So on to the agenda for today. The plan is to start with 1 hour of presentation material, followed by about an hour of Q&A, for a total of 2 hours. Charlie will start us off with Pure's platform strategy and the Enterprise Data Cloud, key enablers for us moving forward. Rob will then talk about Pure's fundamental advantages, which, frankly, are still not super-well understood on the street. Then Prakash will talk about our customer journey to Pure's Cloud Operating Model, which is also an area that's not super-clearly understood, and we hope to increase that understanding here today. And then Bill will wrap us up with Pure's Hyperscaler business. I hear there's just a little bit of interest in that. So we thought we'd cover that.

So then as Charlie comes up here to start us off, I'm going to remind you all, as I've been known to do, that at Pure Storage, it's all about one thing, the software, the software. You're going to hear that all day today.

Charlie?

Thanks, Paul. And thank you all. Good morning. Thank you all for coming. I appreciate you spending time with us this morning. So yes, I'm going to be speaking to you about our overall strategy and how it all fits together. And of course, we'll be speaking about this. This is substantially the same presentation that I'll be giving this afternoon to our New York customers. There's a lot of similarities to the presentation I gave just 3 months ago at our Las Vegas Accelerate conference.

But I'm going to start off by just saying that we continue to outperform the market. Our expectation is that we will continue to do so. We had a very strong quarter this past quarter, as you well know. But something that I think probably is -- well, the milestone that I saw in this last quarter was something that if you dig down into the financial metrics of us and our competitors is probably something that might be a bit surprising. We are now spending as much on R&D as any one of our competitors regardless of size, if not more. We're spending more than most of them, and we're now at the same level as many of them. And so -- and of course, since we spend it just on one software environment to cover all of the different areas of storage, we can be very efficient and have a lot of leverage in that investment unlike some of our competitors. So we think this is a major milestone as we continue to scale in this area. There's really no opportunity in the data storage market that we can't go after at this point.

If we look at the metrics of the company, of course, 13% year-over-year growth, strongly profitable this last quarter. In addition to that, now, as you know, we're in the 40s and probably will be through the year in terms of the percentage revenue in our subscriptions. And as I mentioned, not only are we continuing to be the company that leads in terms of the percentage of our revenue we spend on R&D, but now also in terms of the total dollars.

So the question might be, well, what is spending on R&D give you? Well, one of the things that gets you consistently and now 11 years in a row, top of the charts in terms of the Gartner Magic Quadrant. The other thing is very, very high customer satisfaction. As you know, we've always led in terms of Net Promoter Score, and I don't want to stand in front of the number there. We've always led, and we continue to lead. It's consistently above 80. And it's not just based on the customer service that we provide, but frankly, it's on the quality of the products and the quality of the customer experience with our company and with our products.

So now I'm going to switch gears a little bit and start talking about what is it really about our products and the direction that we're going in that we think is going to impress our customers the most. And I might start from a surprising point in time because I want to address the elephant in the room. So what is the elephant in the room? Well, it's AI, okay? And this may -- so bear with me. The thing about AI that we think is really going to be driving customers as we go forward is the fact that it changes the relationship between software and data. And before AI, you may remember a famous individual once said that software was eating the world. right? And software has really been the driver of the fortunes of tech companies and the driver, frankly, of productivity in business overall.

But what's happening now is that data is starting to drive investment and certainly starting to drive the performance advantages of some companies over another. So maybe data is starting to eat the world and maybe even data, and this has been discussed by many of you, is eating software, putting software as a dominant element somewhat at risk. So it's really changing the nature of software and data. And of course, data is where we play.

So how do we help in this environment? Well, let's step back for a second. Let's talk about enterprise data. What do I mean by enterprise data? I don't mean sort of the files that you all share with your colleagues in your business. I'm talking about the data that business runs on. So the databases, the unstructured data that corporations collect to be able to get a better understanding of their sales force, of their customers and so forth.

So let's talk about how they manage that. They manage that, of course, in data centers. And how are these data centers constructed? Well, let's take a typical application, which might be a database application. The database application starts with the software and the database and the compute that the database is going to be running on. Well, then what they do, what an IT organization will do, is add storage, add networking, and it becomes a stack. And sometimes they call it a full stack. And the full stack is made up, as I said, of compute, switches and storage. And then there's another application. It could be e-mail, it could be just a customer database, it could be an understanding -- it could be their marketing database, and they set that up. Or it could be their virtualization environment. And each one of these is set up.

Now a more sophisticated customer will have a standard for their compute. They may have virtualization. It's all virtualization standard so they can move their application around between different computers. And of course, they're usually standardized on one network environment. But it turns out that they choose their storage bespoke to the application, high performance, low cost block or file. And each one of these can be quite different, as you can see in the picture.

So what does that really mean? What is that -- what is the effect of having this kind of architecture, vertical stacks and bespoke storage? Well, what it means is, first of all, the storage is provisioned individually and manually. It doesn't operate the same across all of these different stacks. What it means is that they can't share capacity from one array to another. Each array is unique to its environment. If it runs out of capacity, even if there's spare capacity somewhere else, it doesn't matter. They have to add capacity to that one array. It means that new services have to be configured manually on every array. It means that any kind of policy that the company sets for their data as a whole has to be done individually array by array. So your data governance standards are also implemented manually.

Now let's think about it, not from the storage angle. Let's think about it from the data angle, how customers manage their data environment. So again, within the same architecture, data is now captive to the application stack. You have to go through the application to get access to the data. The data is not available otherwise. It also means that there are poor records. People can easily copy and paste data in different areas, but the records of that, again, if they're kept at all, it's manually kept. So think of it from a governance standpoint, right? It creates a mess, hard to follow. Raw data is inaccessible to applications such as AI, has to be copied to an AI-specific data storage array somewhere. And then finally, if you think about data governance, because data governance is done array by array and manually, it's going to be inconsistent because this is spread out all over a large corporation, different users, different administrators. So it's very inconsistent data management.

Now let's contrast it. That's not the way that clouds operate. It's not the way that clouds are built. How are clouds built? Very different philosophy overall. They're built largely horizontally. They want everything to be the same across their cloud, not only in one data center, but across data centers. So it generally starts with compute, one compute layer. The compute layer is treated as a general purpose compute layer. That's all tied together, of course, with the network.

Now the next thing to think about is what they do with data storage. The data storage is the same horizontally across the entire organization -- across the entire cloud. The only difference is they may have different price performance layer. They have a high-performance tier, they have a low-performance tier, usually a tier or 2 in the middle. And they'll even have, in many cases, an archive tier, which might be tape.

And so what they're able to do here is any application can access any portion of the storage. Of course, having -- you need to have the right authorization to be able to do so, but there's no physical limitation to what storage can be used for what purpose, right? So it's a completely shared environment that is software defined. It's a management system that creates the environment that determines how to -- or what storage can get access, much more flexible architecture. So this is what we feel is allowing an enterprise storage architecture now to be changed.

An enterprise storage architecture as it exists today is vertical, right, vertical stacks. A cloud storage architecture is horizontal with everything being the same across the horizontal limit. An enterprise storage architecture is largely manual. It is not orchestrated by software. Cloud storage architecture is fully automated and standardized. Enterprise creates data silos, cloud storage architecture creates accessible data pools. And then finally, enterprise storage architecture is very physical, with physical connections. Cloud storage architecture is virtual. It's all based on software, what gets connected to what. So we believe this sets up for the industry a new vision for data, how data will be handled inside the enterprise. And in particular, as you might be able to guess now, it's a cloud -- we want to bring a cloud operating model to enterprise storage.

So how do we do that? Well, we've been working on this for many years. As many of you know, it starts with Purity. Purity has this -- it's a common operating environment for any type of storage with our, very famous now, Evergreen architecture, which means it never gets old. That now -- Purity we've designed -- we started with block, but now provides for block, file and object. In addition to that, it supports all types of data, including everything from the highest performance in AI, all the way down to low cost, now replacing all disk environments. And that we continue with our progress about hitting ever lower price levels for low-performance storage.

We also provide all kinds of services, and that's all embedded within the Purity environment, and we continue to add more and more services to that, all managed by -- sorry, let me rephrase this. This is what we mean by a unified data plane. That is one software environment that covers every level of price performance as well as all different types of data. Now it is all managed by one management system, which we call Pure1, and it is able to be acquired either as product or as a service with our Evergreen//One management system.

Now the big reveal, if you will, that we announced in Las Vegas, and we are going to continue to focus on is this capability that we call Fusion, which now creates a unified data plane. So we're able to support all kinds of enterprise apps, all kinds of modern apps. So there's really no application environment that our system -- that Purity doesn't support today.

But with this addition of what we call an intelligent control plane, instead of our arrays operating as individual arrays, they operate as a cloud of data inside our customers' environment. So all of the arrays communicate with one another -- are able to communicate with one another and are able to operate as a cloud rather than as individual arrays. And this fundamentally changes the way that customers can access their data, use their data and manage their systems.

And when we say that the arrays are networked operate as a cloud, we're talking about a global data cloud, not just within an individual data center. So -- and this is what we mean by an enterprise data cloud, which is the architecture that our system now, with Fusion, allows our customers to build inside their enterprise environment.

So just to sum this up, we're taking arrays from being manually provisioned to being auto provisioned. We're going from dedicated capacity per application stack to automatic load balancing across an entire global network. We're going from governance being manual to global governance of the way data is managed through software control, true software definition of the way data is managed. We're going from policy setting to -- from being manual to being automatic, which means protection is not applied array by array, but by policy. And then finally, from a captive data -- from data being captive to an application stack to data being globally accessible for things such as analytics, such as AI.

Or another way to think about it is the Enterprise Data Cloud allows organizations to make their production -- turn their production data into their data lake. Today, data lakes have to be just a separate pool of storage. Why is that? They already have the data. It's already sitting in a production environment. It's already real time. Why not have that be their data lake. And so it replaces siloed data, if you will, with a data cloud. So you remember, I started this with AI. The opportunity for companies -- one of their biggest challenges is their data is spread out everywhere and siloed. And AI is having companies relook at the way their data architectures exist, and we think this is what sets us up for really a fundamental transformation of architecture inside the enterprise.

Now I want to speak about one more thing in this area, and that is -- I want to speak a little bit about our launches. And I'm not going to go -- this is -- you're going to see these are the launches that we're putting out just this year. This is also what our investment in R&D is bringing to the market now. And I think you'll see that we have an extensive set of announcements that we made and products that we're bringing to market that cover everything from unifying even more the data plane so we can cover more use cases, more tiers, if you will, of storage, but also the intelligent control plane that allows customers to more easily manage this -- not only manage our products, but frankly, to manage their data sets overall.

So I'm not going to go into this, but I do want to touch upon one thing. Again, it's another area that I know there's a lot of interest in, and that is FlashBlade//EXA. So -- and the reason why I want to touch upon it is because -- and Rob will go into much more detail, but it shows how extensible and how flexible our overall architecture inside the product truly is. So EXA is based on Purity, and it's based on our FlashBlade physical architecture. FlashBlade is our scale-out architecture. And up until now, we've used FlashBlade as both a storage target as well as the product that is able to handle the metadata that allows applications to get access to that storage.

What we've done is we've leveraged the fact that FlashBlade is actually the world's fastest metadata engine. And now by attaching additional data nodes to FlashBlade, rather than using FlashBlade as the data store, we now just use it as the metadata engine. And not only can we add sort of unlimited amounts of additional data at high speed to it, but we can replicate the -- and we can expand the metadata engine, and we could just keep expanding it and expanding it.

And what's really magic about this is -- and this is something that every other vendor would love to have is it scales linearly. Every time we add a new FlashBlade metadata EXA engine, it just continues to scale the overall performance, all the way up to right now, internal testing, 10 terabytes per second of read, 5 terabytes per second of write, roughly 5x faster than anything that's out there today. So we're able to cover the entire range of AI needs from hyperscalers, which is going to be based on direct flash, not on EXA. And that will be covered a little bit by Bill today. FlashBlade//EXA, which covers really the GPU clouds, Rob will be covering that. And then regular FlashBlade, which is -- perfectly, is a great product and covers really the entire needs of most enterprises in the AI environment. So we're very well covered. And all of these -- well, that is EXA, S, as is true of all of Pure's products, are covered by Evergreen and allow for the ability to be able to do upgrades in with -- allowing customers to save the investment that they've placed in us.

So summing up, the Enterprise Data Cloud now is really a new architecture for our customers to think about how they virtualize their data environment. It does a number of things. It creates a virtual cloud of data, but it creates a very consistent environment for all of their different workloads. It allows for data to be governed by policy and software rather than manually through fingers on keyboards. We build in cyber resilience. And, again, that can be done by policy rather than fingers on keyboards. And it is completely software-defined. So this is a really fundamental architectural shift in enterprise. It's one of the -- it's both an opportunity for us as well as a challenge because changing architectures in organizations that have been doing the same thing, the same way for decades is not for the faint of heart, but we think we're making great progress here.

So with that, I would like to turn the stage over to Rob Lee.

Thanks, Charlie. All right. Good morning, everyone. Great to see everybody here in person. As Paul mentioned, I'm going to spend my time today talking about and going through Pure's fundamental technology advantages. And what I'm not going to do is actually go through product by product or unpack the portfolio. Instead, what I want to focus the discussion on is really looking at the various pieces of differentiated technology, IP, intellectual property, we've created across the entire storage stack.

We'll then transition into looking at, hey, how do we go take those core blocks of IP, leverage them in different ways, package them in different ways to go meet the needs of different markets and pursue different market opportunities. And then we'll do a double-click into one particular area, which Charlie mentioned, which is the performance and scalability of our metadata, which we see really driving the success of our AI platforms.

If you're out there and you've been following us and you're perhaps still wondering, hey, what's that one critical technology advantage that Pure has that sets you apart from the others? They're in storage, they're an outlier. They do Flash, others do Flash. There's something different up here. if you're out there still wondering what that is, if I've done my job right at the end of this, a, you'll have the answer to that question. And b, you'll walk away realizing it's not one thing. There are 4 or 5 different things that really come together that we're able to now put together, combine in different ways to go after the various market opportunities that we're pursuing as a company.

All right. So we're going to go deep. Before I do, I want to set some context and really just step back and look at, hey, what's in a storage system, a storage system, any storage system out there, one that we might build, one that our competitors might build, one that hyperscalers might build for themselves? And at the most basic level, any storage system has to solve a lot of the same needs. You need physical media. You need somewhere -- something to write your data down on, whether it's hard disk drives, whether it's flash. You're going to need some software to make that physical media work, whether that's OS drivers, whether it's firmware, whether it's purpose-built software. You're going to need some system hardware, right? You got to -- you need chassis, you need sheet metal, you need power supplies, controllers to put all this stuff in and to run your software on.

Once you've assembled all that, you probably want some way to organize, store, retrieve your data. You want all the protocols, performance levels. You probably would like the system to be resilient to faults. You'd probably like some protection, snapshots, replication, that sort of thing. And at some level, you're going to need a way to manage, monitor and administer this thing, right? At a 60,000-foot level view, all storage systems have a very similar set of needs. Now the thing is as you dive down to deeper altitudes, as you come down to 30,000-foot level, as you come down to ground level, even though the needs are very similar across the board, hopefully, what you'll see through this discussion is the way that those needs are addressed couldn't be more different in our industry.

All right. So let's walk through each of these layers and take a look at how the rest of the industry has tried to approach these and what Pure has done differently, right? And we'll start with the physical media and really the software approach that's been taken to control the physical media on which storage is placed. And this is perhaps the most readily apparent difference that we have relative to the competitive set. Everybody else out there is reliant on SSDs to consume flash.

So if we take a step back, what is an SSD, right? An SSD is a technology coping mechanism to make the physics of flash work in a world that is built with software that's designed for hard disk drives. Flash behaves very differently. All the software in the world was built, prior to us, with hard disk drives in mind. So the industry came up with the SSD as a coping mechanism, as a translation mechanism to make that software work for flash.

Now what does that mean? It means that the SSD has a tremendous job that it has to perform. It has to do a ton of work internally to translate flash into a world where it behaves and looks like a hard disk drive. It has to manage the flash at a physical level. It has to perform a lot of remapping, a lot of complex translation of instructions. It has to do background work like garbage collection. It's really complex software. And SSD is basically like a little computer, right? It's got an ASIC in it. It's got a processor. It's got DRAM. It's got complex software. They just call it firmware. It's basically a computer.

The challenge with that is really twofold. One is, as with anything, if you add translation layers, if you add a man in the middle, so to speak, you're fundamentally adding inefficiency. You're adding extra work to be done. That extra work comes at a cost. That cost can be measured in components. It can be measured in performance. It can be measured in reliability. It can be measured in limitations. The second challenge with this architecture is that as flash advances, as the industry builds denser and denser flash, it becomes harder to work with. As you try to put more flash into the drive, you're complicating that task even further. The SSD architecture has reached its limits.

So what has Pure done? Well, we had the benefit of not being saddled with retrofit architectures. We did not have software built for hard disk drives. So we started our software top to bottom, built specifically for flash. What this has allowed us to do is to build a media software layer -- layer of software to manage the physical flash that doesn't add all this translation work, right? We simply avoid this duplicative work by designing a software layer that allows us to expose the best properties of flash directly to the software that's running above.

This also has another side effect, which Bill will talk to in a bit in his discussion, which is that as we look at different ways of packaging the software, we're now taking on the task of making all the different types of flash and NAND out there in the world behave and look the same. For everybody else that's relying on SSDs, well, the SSD manufacturers have put that work of qualification, put that work of the different behaviors of firmware on the consumer, whether that's the enterprise OEM vendor or whether it's the hyperscaler.

All right. So let's move up the stack a little bit. Let's take a look at the system hardware. These are the chassis, the enclosures, the power supplies. And this is an area where our competitors, frankly, treat this as an afterthought, right? You can look at this, you can tell this by -- based on what they ship, right? It's largely third-party OEM hardware, it's repurposed servers. One of our competitors, at their top-of-line product, if you look at their manual, will say things like, when you install a storage array, you have to leave extra rack space above and below this thing to allow for sufficient airflow to cool the thing so it doesn't overheat and to allow us to go and service it. That's how much care and thought they have put into the system hardware.

Obviously, that results in complexity, bloat, operating costs. We've taken a different approach, right? In our world, our system hardware is integral to what we do, right? It's integral to the Evergreen promise. Evergreen at its most basic level is a promise that once you're on Pure technology, you're on a path to non-obsolescence, right? You have the ability to non-disruptively evolve, upgrade scale to future needs, whether they're -- it's future technology, whether it's more capacity, more performance.

Well, the only way we can go and deliver that is a hardware platform that has the capability to be forward and backwards compatible that's designed for decades-long lifetime. Our system hardware is core and integral to what we do. And also by owning that system and really investing heavily in it, it allows us to create significant advantages that come through in reliability, performance and simplicity.

All right. Let's move up the stack one click. Take a look at the main layer of storage software, right? And this is where I would say some of the biggest differences really start to come out. If I look at the competitive set, this is where the retrofit architectures rear their ugly head yet again. Most of our competitors are building on retrofit software stacks designed in an era of hard drives. They were designed in an era where hard drives are basically the same speed they are today when flash has gotten 100 to 1,000x faster. They were designed in an era where networking was 100x slower. They were designed in an era where CPUs and processing was 10x slower.

As you might imagine, right, the choices in how you build software, how you architect, how you design software in that world are very different than how you would design in a world of hardware resources that we have today. I'll give you a very simple example. If you're designing software to place storage on a hard disk drive, it's a physical moving platter. If you want to optimize for that, you want to lay out your data sequentially because it's a sequential mechanism, you want to lay out your data sequentially, so as the platter moves around, you very easily read it.

Well, in a world of flash where it's highly parallel, it's semiconductor driven, if I want to go and optimize that, I want to place my data on multiple chips, and I want to go and access those entirely in parallel. These are very orthogonal decisions. This is one of many examples of orthogonal design decisions that are central to software stacks that are designed either for the hard disk era or for the flash era. As you might imagine, these are incredibly hard to unwind.

And one of the other offshoots of the software that was designed in the era of hard drives is that because performance was so difficult to achieve, the competitive set generally had to build different software stacks for different purposes. Charlie mentioned this before in terms of how different storage arrays have historically been provisioned for different applications. Well, the competitive set generally has had to build separate products and software for each different need to meet that performance. Those of you who cover software will realize that it's incredibly hard once you have that fragmentation to bring it all back together.

So what has Pure done? Well, again, we have not been saddled with. We did not grow up in an era of retrofit software. We were able to design for flash since day 1. We were able to build from the ground up for the speed and scale that flash was capable of. And we made those hard investments to really take advantage of not just the performance that was available to us back in 2009, but where we knew flash was going. More importantly, we've very, very intentionally invested in keeping a unified and single software base and single hardware technology.

This now brings us back into -- and is a key enabler for Evergreen as well, right? If you think about the core Evergreen promise, which is non-obsolescence is the ability to non-disruptively grow, scale, evolve in different ways. Well, how do you do that in a world where you have 6 different software stacks to meet different performance levels? It's really, really difficult, right? One of the things that is critical to enabling Evergreen, besides the system hardware, is the flexibility and the dynamic range that our software allows us -- our single software base allows us to meet the entire set of needs from AI to archive.

All right. Click forward. Thank you. All right. So let's take a look at management and monitoring. Again, this is an area where if I look at the competitors, generally thought of as an afterthought. It's a bolt-on. Frankly, most of our competitors' management tools are more focused on all the tuning knobs and configurations that are required to make those products perform. If you look at how we treat management and monitoring, it's central to not just how we deliver a customer experience, but also to how we build a full as-a-service model.

Let me step back and unpack that. If you look at all of our arrays since day 1, right, the Phone Home data to Pure1, the intelligence that we can gather from the arrays, we manage -- if your home thermostats, if your AV system at home is able to be cloud managed and that service is able to be improved over the air, why can we not do that in the enterprise? That's been our focus. That has allowed us to have the NPS score, the customer satisfaction, the proactive support experience that we've been known for.

That also allows us to better understand the arrays in time and transition that into a full as-a-service offering. You heard Charlie talk about the intelligent automation. Prakash is going to go into this quite a bit more. All of this stems from our investment in the management and monitoring part of the stack. All right. So we've unpacked.

Let's move forward. Thank you. All right. So we've walked through each of the layers of the stack. What I want to transition us to now is looking at how do we put these pieces together, how do we leverage the investments we've made in this differentiated technology at the different levels to meet the needs of different markets. And the 4 markets that we want to focus on are: number one, the core enterprise; I want to take a look at what we're doing on the cloud and cloud-native workloads; the scale AI, which is kind of more the neoclouds, the GPU clouds; and then certainly take a look at what we're doing and how these pieces fit together for the hyperscalers.

So the core enterprise workloads, and this is all the workloads you'd find in an enterprise data center. This has been the North Star. This is where all the pieces come together, and it makes sense, right? Our enterprise clients aren't coming to us with a partial solution. They want us to provide the whole kit and caboodle, right? This is where by tightly integrating all of the layers of the stack, the differentiation we create at each of these layers, we're able to create a very compelling value proposition across all these layers. And you see this coming out in our flagship products like FlashArray and FlashBlade.

If we shift a little bit and talk -- and take a look at what we're doing for customers who perhaps are running on top of public clouds or running cloud-native workloads on top of third-party infrastructure, this is where, in that case, the public cloud -- the CSP might be providing the physical infrastructure. They're providing the physical media, they're providing the system hardware. But customers are seeing value in having the storage services, the same enterprise capabilities, the data protection, the reliability, the sameness of management across on-prem, hybrid and cloud. And you see this model coming out and these areas of IP coming together in products such as CloudBlockStore or Portworx.

Shifting gears yet again. If we look at the needs of the scale AI community, the neoclouds, the GPU clouds, the largest foundation model builders, these are folks that have their own higher-level software management, monitoring. They have their own higher-level data flows. In a lot of cases, they're specifying their own physical hardware, ultimately kind of optimized for high performance. What they value, what they get from us is the storage software. The storage software that gives them that scalability that unlocks the performance of the hardware underneath us but also allows us to fit nicely into their software ecosystem, their management, monitoring systems.

All right. And then last but not least, hyperscalers. I'm not going to go super deep here because Bill is going to come up in a few minutes and unpack this further. But this is where -- if we step back and we look at the hyperscalers, they've got their own management, monitoring. They think about storage reliability and storage services at data center -- across data center scale. They design their own system hardware, but they've got a challenge, right? They've got a need to move to flash. They're largely running on hard disk drives today. The SSD architecture, I've already explained, has significant limitations.

We have the capability. We have the means in our DirectFlash software to dramatically improve their path. And the means are really driven by the software, right? There's nothing -- our DirectFlash Modules, there's nothing -- there's no secret sauce in there. It's -- the enabler there isn't how we solder chips to a board. The enabler there is what we have put into the software to make that work. And that's really key to the integration that's driving the Hyperscaler business, which, again, Bill will unpack further in a few minutes.

All right. So if we step back, hopefully, this helps illustrate how -- starting with the core enterprise business where we focus on building a tight integration between all these layers, coming together to really meet the full needs of an enterprise across their entire set of applications has led us down the path of building core differentiated IP blocks. We can then go take those IP blocks and leverage them in different ways to meet the needs of other markets, whether that's folks that are running on cloud, whether it's cloud-native applications, whether it's the neoclouds and scale AI or whether it's the hyperscalers.

I mentioned I wanted to do a double-click into one area, please. Thank you -- which is the scalability of the metadata. Charlie mentioned this upfront. I want to unpack this a little bit. We talk about metadata, we talk about data. What does that really mean? So in its most simplest terms, imagine you're trying to load a file. You're trying to load a file and transfer it somewhere. There's the work of looking at the file by name, figuring out what directory it lives in, figuring out, do you have the permission to open it, figuring out, okay, once you've identified that file, where are all the bits of the file located, on flash or on disk.

And then there's the work of actually reading and transferring the file. There's the administrative work upfront, the metadata and then there's the actual data transfer work at the back. Well, it turns out both are important, as you might imagine. And it turns out that the metadata, the administrative work, scaling that, making that performance level super high is actually, in many ways, more challenging than scaling the data transfer performance. It's more challenging.

And one of the reasons it's more challenging is you've got to figure out a way to organize that metadata, keep track of all those bits and pieces in a way that's highly optimized for parallel access. You've got to keep all that information in a way that makes it easy for you to scale and add performance into the system. And the second challenge you have is once you've organized your metadata in that way, you've got to optimize the coordination. You've got to allow that data to be accessed in parallel without the different processes stepping on each other, without creating wrong results, without creating corruptions, without creating inconsistencies. And so these are fundamental challenges.

How have we solved these? Well, it turns out that we have focused in this area since day 1. I mentioned before, we were not saddled with hard disk drive-based software. We focused on the parallelism of flash since day 1. One of the key enablers that has allowed us to do that is how we organize metadata in the system, not just file metadata, but our internal metadata. We've organized it in what's called a key-value store. Frankly, we've borrowed concepts that come from the database community. We've borrowed concepts such as key-value stores, such as distributed transaction engines. If you open up our software and look under the hood, it actually looks a lot more like a high-frequency trading system or a distributed database than a traditional software stack.

What this means is that the scalability of the metadata performance we can deliver, all of that administrative work to pair with the data transfer work is extremely high. We see these needs as super acute in technical computing, whether it's chip design, whether it's fluid dynamics or whether it's AI. And this has really been the driver of strength for the entire FlashBlade portfolio since really its inception, whether it's FlashBlade//S for the majority of the enterprise high-performance needs or, as Charlie mentioned, with FlashBlade//EXA, where we're keeping that strong kind of scalable high-performance metadata core and then pairing it with an open hardware architecture to meet the needs of some of the world's largest training and high-performance environments.

And as Charlie mentioned upfront, this now completes our portfolio of serving the needs of AI at the entire range of scales from the smallest research groups to what the enterprise is doing, which is largely served with FlashBlade//S, your neoclouds, your tech titans, what we're calling the scale AI community with FlashBlade//EXA or what the hyperscalers are doing, which is really more of a horizontal high-performance design, which we're serving with DirectFlash, which again, Bill will go into greater detail on.

So with that, I'm out of time. I'm going to turn the stage over to Prakash, who's going to walk us through how we're extending the management and monitoring plan to deliver on the Enterprise Data Cloud.

Thanks. So as we think about our customers, our goal is to help them move towards this cloud operating model. Charlie unpacked what an intelligent control plane is. But if you think about Salesforce or any SaaS solution on the market, realistically, how they're able to get economies of scale and SaaS margins is by standardizing operations. When you run everything, when you control everything as a SaaS-managed thing, your environment is very standardized. But if you go into a customer data center, it's not. Charlie mentioned it's very siloed. It's very fragmented.

And for anyone who's been in a data center with a number of solutions and the number of operating systems that have historically existed, it's a mess. If you want to start moving people to an Enterprise Data Cloud and get SaaS economics, you need to standardize the components, create standardization at scale. And what that allows you to do is provide instantaneous value where you can spin things up at the need of what your requirements are.

So our cloud operating model is driven to create the standardization. As I unpack this management and monitoring, historically, people thought about this box-to-box storage management. It's a capability where it's like I need this type of data, I need to provision it, block file systems volumes, et cetera. That paradigm to move to a SaaS operating model requires a paradigm shift. You can't think about these low-level objects and management anymore. You have to think about how do you move to policies and fleets. You need fewer components that are more standard with simplified policies where you can move beyond low-level management to higher-level management.

Now this is fundamental in the shift we're driving in our customer and install base. We're the only vendor that's driving this innovation to this policy-centric view. Now that's a step in the direction. But when you move beyond these policy and management objects, you then get into the world of what AI needs, which is what are my workloads, what are my workflows.

So managing data sets is the ability to take a look and assign these policies and these management principles to workloads. You want all of your SQL workloads to operate this way. You want all of your Tier 1 applications to operate that way. And all of these policies create standardization in a data center. It's the reason why a lot of hyperscalers who've tried to do go on-premise have failed because they expected standardization in a world where there was none. So you have to take people along this journey.

Now once you get there, there's 2 approaches. You could do this. We can provide the capabilities with our management and monitoring for customer teams to deliver customer SLAs. Or secondarily, if you don't want to worry about it, why not trust us as the vendor to deliver vendor-managed SLAs? So when we talk about our Evergreen//One offering, it is a vendor-managed SLA. It's a vendor-managed service with standardized building blocks.

You can get whatever storage you need, block file or object, any type. You can get it deployed wherever you need it in your data center. When we run our service, think about it this way, we're paying for the power and rack space we use. So we're just running our service and saying, "Oh, this is our service. It's a managed endpoint. We're running it in your data center. With our CloudBlockStore, you can get it in the cloud. We have partnerships with MSPs where you can get it in the Co-Lo, and it's available in 32 countries around the world."

But the standardization, when we talk about that standardized operating model at scale, what do you want to standardize? Charlie mentioned performance levels. Without a single operating system, you couldn't get unified performance and actually deliver this at any reasonable margins. If you want uptime, 0 data loss, like we've had a lot of people think about data center consolidations right now. We all know about the power crunch. We have a site relocation SLA where we can help customers consolidate their data centers, their physical locations because our service can deploy new hardware and move the data seamlessly transparent to the customer as part of the site relocation service.

With our Evergreen architecture, we have 0 planned downtime. We have guaranteed watts per terabyte with our energy efficiency guarantees and cyber recovery SLAs in case of ransomware attack, where we guarantee the time it takes to get up and running for customers based on their data sets. So we were unique in approaching this problem set, not from just a financial lens of cash or credit, how do you want to move to recurring revenue, but fundamentally changing the operating principles of how do you standardize data centers with this cloud operating model.

Now it's built on the uniqueness that we talked about. You could not do this without Evergreen because a fundamental approach of a SaaS solution is for the same unit economics, your offering gets better over time. You expect -- if you're paying a subscription to Netflix, you expect more content. If you're paying a subscription to a storage system, you need to ensure it gets better over time. The Evergreen architecture allows us to have always upgradable hardware. And this intelligent control plane is fundamentally changing the game by linking together 2 disparate spaces, observability, where you collect data; and automation, where you can actually take action on the data you collect. So let me drill into this paradigm shift because I think this is very important for folks to understand.

Our -- historically, we've collected about 70 petabytes of data in Pure1 on an annual basis, like so over the last year, we know everything from how people -- like how the system operates, is low-level NAND, Bill will get into some of that. But we also know how people use storage. Think about Tesla, there are advantages in miles driven on the road for self-driving. Our advantage over 15 years is we've collected more information on how people use flash than anyone on the planet.

And if you take that information and you combine it with Fusion, where our arrays are now networked. Charlie mentioned the network effect of arrays, you can go ahead and actually take action on different things, saying, "Hey, your environment is unbalanced. Let's dynamically rebalance your environment." So you're completely creating standard building blocks and optimizing the landscape.

So this paradigm shift allows customers to manage fleets, manage sites, these logical sites, have policy-driven workload management, whether it be protection, tiering, et cetera. And this -- the ability to do it continuously is like a CI/CD pipeline for developers. It's not onetime, it's not manual, it's continuous. So our hardware has been getting better over time. Now your operations are getting better over time. And that's the power of the intelligent control plane.

Now this Evergreen//One offering we talked about, you get block, file, object, on-premises, cloud, Co-Lo and all these SLAs, are built on this concept of this intelligent control plane because we need to continuously think you're running a service, you have SREs running the service. You need to be able to go ahead and have the capabilities to say, "You know what? If you want the service to run, operate efficiently, how do I change things dynamically in the environment?" You need this approach to say, "Let's keep the site rebalanced. Let's ensure that we can offer these ransomware recovery SLAs through policy." So as Rob talked about the core bits of IP, our service offering to deliver vendor-managed SLAs is built on this management and monitoring paradigm with this differentiated intelligent control plane.

Now let's think about this competitively. We invented this Evergreen architecture as a core fundamental building block early in our company's history. This wasn't a bolt-on. It wasn't an after effect. But our Evergreen//One offering now allows us at the end of a service life to continually just say, "Oh, we can keep this hardware upgradable and current." It gets better over time. We guarantee in our contract, you will never have a data migration once you move to Pure. That's part of this offering.

Most of our competitors at the end of the service life cycle contract is like, "Okay, now you have to pay for migrations," or they'll have to subsidize that in their unit economics to migrate from one generation of technology to another.

Our service definition isn't starting with a box and a lease. We define these standard building blocks and SLAs that are set up by site. Here's a service site. This is what I need at that site. I need this performance tier and this capacity tier. We have a simple contract. All of our SLAs are transparent. It isn't like, "Oh, go," all those terrible warranty claims that you have to go and read, exclusions, et cetera. Most of our competitors have exclusions for planned downtime because they don't have this Evergreen hardware. They have exclusions for data migrations. All of these exclusions exist very similar to -- I think I said this to you guys last time, it's like all the fine print on a drunk commercial.

Transparent SLAs. An SLA has real teeth because you have real financial penalties, and customers can monitor and see it in your product. Our management and monitoring in the product shows you any SLA violations. They're triggered right from our product. And then finally, because we're running a service, we pay for the capabilities that we're doing. We're renting space in your data center, so we'll pay for it with rack space and cooling. So these are some of the unique advantages that make Evergreen//One unique.

And while we see people on this journey to these SLAs, some customers prefer to do it themselves. So with our intelligent control plane, customers have large service delivery capabilities, and they're building customer-managed SLAs. But increasingly, we're seeing strong adoption where customers prefer, "You know what? I'm not going to be able to standardize. I've got too much variability in my environment. I'm going to trust you as a vendor to offer vendor-managed SLAs in my environment." And we've now crossed over 1,000 customers in Evergreen//One in just about 4 years. So this offering is now starting to see broad adoption across a wide variety of sectors.

Now this Enterprise Data Cloud creates a network effect because once you're on this platform and once you have these policies and once you have the standardization, it's a one-way street. This unified data plane gives you a single operating environment that you can program with our intelligent control plane right into the automation and build in your business. When you can provision things, you get built into customers' workloads and workflows, which creates a platform effect and stickiness for Pure as a vendor.

So this allows us to make sure the next purchase that a customer thinks about isn't a jump ball. It's not subject to just NAND economics and dollar per terabyte because your value is critical to a customer's workflow. And if you think about getting built into a customer landscape, how do you think about vendors who do that well? Salesforce did that with Force.com. It was a platform where you could build extensions right into the environment. So if you think about platform companies, platform companies are unique in their ability to get embedded directly into the core operating environment of a customer's core IP that's revenue generating or cost optimizing. You're driving integration directly into their businesses. This platform effect with this Enterprise Data Cloud allows us getting built into real customer workflows and become core to standardizing their operating environment.

So with that, I'm going to summarize and talk about this paradigm shift. Our vendors keep -- our competitors keep talking about this world, storage management. And some of them are even like, "Oh, I'm going to package up database management software with this stuff." But they're fundamentally missing this paradigm shift around how do you drive standard operating procedures at scale because that's the fundamental thing we need to bring into the data center environment.

Charlie talked about the horizontal nature of unlocking the value. If you don't have standardized components and you don't have standardized operating policies, you're not going to unlock the value of your data. Customers can deliver that themselves, or they can trust us to do it. And increasingly, we're seeing customers trust us to do it.

So with that, I'll summarize, and we'll bring up Bill, and he's going to talk to you about something I'm sure you're all excited to hear about is the Hyperscaler business.

All right. Thank you, everybody. Thank you for your time, coming out today. I'm the last presentation. Hopefully, it will be quick. I know we're getting fatigued. I'm going to talk about our technical advantage, our supply chain advantage and our structural advantage of our hyperscale offering today.

Okay. So first off, I wanted to just step back and talk about what the offering is because sometimes there's a lot of confusion about that. And I talked about it last year, but I just want to go over it one more time, okay? So it's not an enterprise storage array. We're not taking our product and putting it in the hyperscale and just asking them to adapt to our solution, which is something that others have done, and you can make some money doing that. But this is an offering that is much easier for them to consume.

We're taking our DirectFlash Modules and the part of Purity that controls the DirectFlash Modules, bundling that and offering it in a way that's easy for hyperscalers to consume. And it's not a point product. It doesn't offer one service. It can be used across their entire production environment, right? So we take parts of the DFM, which is like a next-generation SSD, the parts that control data placement, management of the NAND, we elevate it to software and offer that as a solution. And we've scaled the solution so that it can be consumed in very high quantities, which is something that's necessary in a hyperscale environment. And this solution has been fully released to production. So it's available today.

So I wanted to talk about the elephant. So Charlie talked about the elephant. And the AI market is affecting this market in a big way. So AI has -- there's a tremendous amount of CapEx spending going on. And it's created this problem in the industry. So GPUs are in allocation. Everybody is fighting to get a GPU. And then you heard about HBM memory. So imagine all these companies spending money to try to get HBM memory to attach to the GPUs, it's created a run on HBM memory. Then it's happened to the DRAM market. Well, this tightness now has hit the storage media market, okay? So there's a tremendous amount of pressure on the hard drive market and on the flash market.

Now the hard drive industry hasn't done much over the past decade or 2 decades to increase capacity, right? So they're sold out. There are 52-week lead times, and they're under a tremendous amount of pressure. This has also come now to the flash market as the hyperscalers are looking for any solution they can get in order to get storage media and increase the size of their estate and address these AI needs. So this has created a tailwind for us. There's a tactical tailwind now where hyperscalers are open to new solutions much more than they were in the past because they need to increase this capacity any way they can. And that's a market condition that we're taking advantage of.

But in addition to a short-term tactical tailwind, we have the long-term structural tailwind of NAND flash versus hard drives. So -- if you think about a hard drive, it's a complicated electromechanical assembly. It's a very mature technology that was first released in the 1950s, and now it's become very mature. The cost efficiencies in hard drives, what they're doing in the factories and how they're improving things is raising the capacity very slowly, and it's lowering the cost per bit very slowly. NAND flash. NAND flash is a semiconductor commodity item, okay? Process improvements, technology improvements in semiconductors, lower the cost very quickly. These innovations have led that over the long term, we see a 2x decrease in NAND prices compared to hard drives. And this is a tailwind that behind the basis of the entire offering. But we don't need NAND prices to become cheaper than hard drives in order for this offering to work because the offering has inherent value. There are many parts of the offering that are just much better than any solution that they can buy today.

So versus hard drives, we're 10x the density. We have 5x less space, lower power, we have fewer failures, and then we last twice as long. And what does this lead to? This leads to a total cost of ownership that's much, much lower for the NAND flash. It also means that the ancillary equipment around the solution is lower with flash. So imagine if you buy a bunch of hard drives, how many controllers, network ports, racks, power supplies, cooling, you have to put around this archaic architecture, right? But against flash, it's much more efficient. So we have this other advantage. So if you look at this graph with all its interesting and precise numbering, the hard drives today are scaling to about 30 terabytes, right? Our module this year, we will announce a 300-terabyte module. It wasn't that long ago that the average module SSD or DFM that you could buy was smaller than a hard drive. And this year, we will be 10x larger for 1 unit. We expect that to continue when we release our 600-terabyte module next year.

All right. So the hard drive market is a very weird place, okay? The global supply of hard drives, 74% of the entire global supply of hard drives are consumed by a small number of companies, think less than 6 companies by all 74% of the hard drives produced worldwide. okay? This is a market that's ripe for disruption. As we work with these hyperscalers, as they move and change their architectures, this is going to have a major effect and immediate effect on the hard drive industry.

All right. Let's talk a little bit about where our solution goes. So in the typical hyperscale infrastructure, they tend to have these horizontal storage pools that Charlie talked about. And they address the data needs of cold, warm and hot tiers. So I think a hot tier is something that's getting accessed all the time. You need very high performance, maybe you need low latency. Warm tier tends to be a combination of SSDs and hard drives and the cold tier tends to be hard drives and sometimes it gets super cold and becomes tape, right? Our solution has a way to span all of these tiers. Because we've elevated the solution to software, we can tune our solution and meet the needs of any of these tiers, different sizes of DFMs, different software tuning, different way to integrate with the hyperscale. And we are the only company that has one solution that can move across all of these tiers.

All right. So we're not just competing with hard drives. We're also competing with SSDs. And as Rob was talking about, we don't really make an SSD. DFM is not an SSD. We've taken the functions that are complicated, that change a lot, that control the NAND flash and moved it to software. So effectively, we've taken something that's found in hardware that's hard to change over time, it's complicated, and we've elevated it to software. And that's made it easier for us to do things like qualify different NAND vendors. But one of the things that I didn't expect when we started this journey was that hyperscalers really value the agility. If you think about what's going on in hyperscale, it's a huge software development environment, dominated by software engineers. They change their infrastructure, meaning their software and how they control storage all the time. And they really value our agility. So SSD companies don't have that system expertise. They don't talk in the same ways that hyperscalers do about the storage stack.

Since in our enterprise solutions, we've done the entire storage stack. That's not what we're offering here, right? But we've done the entire storage stack. So our engineers think at the system level. They can talk about system ideas. So our agility, our ability to change the offering really quickly, offer different protocols, different performance levels, how we handle metadata, how we pass errors forward. It's really been a very strong plus for us over using an SSD. All right. So on the technical advantages, I talked about higher capacity, density, the flexibility of the software. our ability to work with different NAND vendors. Also on the structural advantage, I talked about this tactical win we have right now where there's a lot of froth in the market and people are very open to new solutions, and that's giving us a tactical advantage. But I haven't talked at all about our supply chain advantage.

And when you're purchasing something at very, very high volume, the way the hyperscalers are, supply chain is super, super important. They're not interested in a solution that's hard to buy. They can't get quickly. They can't get a lot of it, okay? So we have signed definitive supply agreements with Kioxia, Micron and Hynix. And those came out in press releases over the past year. Now in the total NAND market, there's only 6 or 7 suppliers of NAND flash, and there's only 5 of high volume. We have 3 of them signed up. If you look at the total NAND market here for each one, how much NAND flash each one make, 15%, 13%, 20%. But all of these are in our solution. So with our solution, we offer to the hyperscalers access to almost half the NAND flash in the world with one solution. We're the only solution like this. If they use an off-the-shelf SSD, they get access to like 15%, 20% of the market. With our -- and this is an environment that's really under a lot of pressure. So we're the only solution in the world that offers access to almost 50% of the market, and that's highly valued by our customers.

All right. Thank you. I'll bring Paul back up to wrap us up.

Thank you, Bill, my hero. So as they bring up the chairs for Q&A, Charlie is going to have just an initial comment about one particular topic.

Yes. Thanks, Paul. So yes, as everybody is coming up, I am going to make an announcement. Of course, we're online so that -- this is an open forum. As you might imagine, as we're working with our hyperscaler customer and hyperscaler prospects, they're very sensitive to their own intellectual property, their own confidential information. And as we've talked more and more to them, they would like us to be as sensitive as they are to having that information out in the wild, if you will. So from this point on, whether it's here in our earnings calls and so forth, we're no longer going to refer to any of our customers or prospects by name, and we're not going to be explicit about any specific numbers relative to those hyperscalers, that is shipments, et cetera. We're working very closely with them. They've made their interests and concerns known, and we're going to respect our customers' intellectual property. So I want to put that out there because I know there'll be a lot of questions here, and we've been a bit perhaps more explicit in the past, and we're going to be more cautious as we go forward with their information.

Okay. Let's jump into Q&A. Let's start with Howard.

Howard Ma with Guggenheim Securities. Thank you for a very informative and tightly packed presentation. Charlie, when I think about the evolution of your product portfolio to what is enterprise data cloud today, I can't help but think about Jensen at NVIDIA talking about a fundamental re-architecture of enterprise data centers. And I believe he said it's something like a 10- to 15-year opportunity. When I think about the advantages of EDC, and I wrote some of these down. So just to say it back to you, unified software management plan, removing data silos, different storage tiers for AI to archive, which I really like that phrase, infinite scalability, multi-cloud, governance, cyber resilience, these make for a highly compelling and I believe, truly differentiated approach.

So when your sales teams and your technical teams go into your enterprise customers today, which I believe one of you called it the North Star, what is the level of resistance versus acceptance of this automated approach that you're delivering? Are we at a -- and sorry for the multipart, but are we at a tipping point, do you think, where the use cases compel this automated approach? And then finally, when you talk about the optimism in your fiscal year guide, when you raise your fiscal year guide, how much of that is due to EDC increasing pipelines?

Yes. Multiparts are okay here.

Yes. It's a great question, and things are moving very rapidly in the market. In terms of your question, what is the resistance, what's the openness? Things are moving very rapidly in this market. Data storage was -- has been traditionally a very slow-moving market. Resistance to any modification of the way that customers bought or even sellers sold data storage was very high. One of the reasons why I opened the way I did is that AI is causing -- regardless, there's been a lot of hype around AI in terms of how much you're going to sell into AI. What -- the benefit of AI for us right now is customers are rethinking their architectures, okay? It's not about how much we sell directly into AI. It's that how open are customers to rethinking their overall architecture in their environment. And so the discussion around enterprise data cloud has really opened up.

Now it is not correct. It would not be correct to say, yes, now a majority of our customers are flocking to this whole idea. But what we are seeing is leading-edge customers really listening and it's starting to affect the way they are thinking about purchasing. Now again, this is at higher levels inside an IT organization, not at the lowest levels. I don't think it's really yet fundamentally changed individual storage admins, although I think it is going to in the near future. But certainly, at the higher levels of an organization -- we are speaking to higher levels in the organization, and we are finally talking to them about franchise opportunities rather than individual use case opportunities. So it's making a difference.

Aaron, please?

Actually, let me just finish.

Sorry, multipart.

And that's what's giving us some optimism about the -- it's not just about optimism. I mean, obviously, we have pipeline analysis in all of the individual financial analysis that we go through when we do forecast, but we are seeing this turn into bigger opportunities.

So let's go to Aaron and then Jason.

Aaron Rakers at Wells Fargo. I appreciate you doing this day. Paul, you should have never said that about multiparts. So I guess my first part of my multipart is, I can appreciate that we don't want to talk about the hyperscaler customers. That's no different than what we've heard from a lot of other vendors that participate in this market. But the comment that you've now released the solution, fully released the solution to production, can you just remind us again on the pace of deployment? I know last quarter, you talked about pilot deployments and importantly, how you see the breadth of these opportunities evolving beyond maybe just that one customer without naming any customers as we look into next year? And then I'll throw my second question out there. Just kind of building on that earlier question, this Fusion layer just Tarek, you're in the stage, how does that get priced? Is that -- is there a software element to what's evolving or deepening in the Pure story that we should all be thinking about here as we move forward?

Let me touch upon that second part first, and then I think between [ Kaz ] and Bill, et cetera, we'll touch upon the first part. So Fusion is not -- we are not monetizing Fusion directly. Fusion now is part of the Purity software. It is -- we call it out because it's a unique new addition to the Purity software, but it's part of Purity software. When we sell software as part of our -- that's embedded in the product, we don't charge extra for it. But I can think of a number of analogies. It creates a network effect.

There had never been -- there has never been in storage a reason why the next array that a customer buys should be the same vendor unless it's economics or a TLA of some type. right? In fact, they often -- every use case, as Prakash mentioned, is another jump ball. It's feature versus price versus whatever. We think this creates a network effect, which is valuable for the customer because it makes their lives easier, it creates more consistency. It will create more efficiency for them. But for us, really, if we reduce the amount of effort that it takes to sell the next array, it will get us larger market share and the larger wallet share.

Yes. So I made a comment in my presentation that we're released to production. So that is a commentary on our status. So about the maturity of our offering, and we've gone through our testing, and we've released our solution to production, which means it can be mass produced. It's not a commentary on the status of where our customer is in their journey. So -- and I can't talk about their status. We continue to have engagements with others. Yes. I can't go into too much detail about what point we are with each one, but we are having conversations, and we continue to work with them.

And we're more convinced than ever that, again, this is the time the transition is starting. And over the next few years, there's no reason for any of the hyperscalers to not do the switch and you saw what it's going to do to the disk market. It's just going to wipe out disks because flash is a better solution that is advancing at a faster rate.

I think another way to think about this is you only have to believe that we are a third alternative to what are 2 alternatives today, hard disks and SSDs. And then the question will be what type of market share out of that do we get? Right now, it's all opportunity for us, right? I mean 2 quarters ago, it was literally zero. So it's all opportunity.

Okay. Let's go to Jason, then we'll come over here to Asiya and then okay, Erik, after that.

Jason Ader with William Blair. Good to see everybody. My question is on the hyperscalers. Can you just walk us through maybe like historically, what the split between HDDs and SSDs were for the hyperscalers? And then fast forward a few years, [ Kaz ] to your point, like how does that mix shift change between HDDs and SSDs? And then what piece of the SSD market do you guys think you can get? And like how -- where do you fit in into that mix because I imagine they're going to continue to use a mix of media.

Yes. So I can talk about that for a minute. All right. So obviously, in the distant past, they were all hard drives. In the -- recently, let's say, within the last 10 years, they moved almost all of that hot tier up at the top to SSDs to TLC-based SSDs. TLC is a version of NAND flash. It's the fastest. And those are -- that is sort of the first layer that many of their accesses go through. So most of that in the hyperscale has already moved, okay? Then you get to the warm and hard drive space. They're starting to integrate more and more technology, including ours, in the warm space and then growing down into the colder and colder spaces. So they are in the transition of doing it today. Does that answer your question?

Yes. I mean I'd probably guess single digits for SSDs as a percentage of exabytes?

Yes. Obviously, it depends on which hyperscaler it is. Some are further on their journey to others, but there are some hyperscalers that are still in single digit on the NAND flash.

You guys are the TLC SSDs, they don't go away.

Right. But our solution can be used in that layer as well.

Okay. But you are actually more in the warm layer, the QLC SSDs.

Warm is where there's a lot more exabytes and that's sort of the best opportunity for us right now. Yes, we want to take the hot layer, but the warm layer is much larger and the economics are just fantastic for us. And then you move down into the whole layer. And if you're a hyperscaler, you're going to start at probably looking at the warm layer because you say, I already have SSDs here. I get less of a boost by switching that. So you're probably going to start looking at the top of the warm layer and again, move down. And as we do a good job for them, move up. Charlie is right that we're never going to have 100% market share. None of these guys will ever bet on a single solution. They always will keep their optionality. Our job is to be the best supplier and get a huge portion of the market share, but we'll never be 100. We'll never be 90%, right? Because they just physically would not ever, ever do that. And so we want to capture every bit that they have, we want to play for, and we want to capture as much of that as possible.

And then Jason, just add one more thing to that. A part of your question was, hey, historically, how much of the mix of deployment in the hyperscalers has been hard disk drives versus flash? I think this is an area where the hyperscalers have almost paradoxically lagged the enterprise in mainstream adoption of flash. They have been able to, through R&D, through just extraordinary means, let's say, be able to squeeze the last bits of optimization out of hard disk drives. We're seeing that start to change. And when we -- when it changes, I believe it's going to happen in a very quick way, right? The hyperscalers get incredible economies of scale through standardization.

They're not going to -- it works against their operating principles and economics to shift gradually over time, 5% here, 5% there over time. They design their data center architectures in generational horizontal designs for a reason once they design that, that now gives them a template to go and replicate and operate with uniformity. And so to Bill's point earlier, the hard disk drive market is an interesting place right now. We clearly have our view on how that plays out over the next couple of years. And I think it plays out in pretty discrete steps.

One other thing I'd add, you asked the question, it opens all the ads. In talking with some of the top engineers at the hyperscalers, they will point out to us how much difficulty they have in using all the capacity on the hard drives, right? As the hard drive grows in capacity, but it doesn't grow in performance, then they need the data to be colder. And so that's another advantage that if you remember Bill's chart, as our capacity goes up and up and up with the flash, they can use all those bits. But if the hard drive vendors came along tomorrow and said, "Hey, we've got a new drive that's 40% larger. The hyperscalers are like, well, I can't use those bits because I'm not getting more performance, and so they lose the economic benefit.

It also means that they'll tell us things like we're overbuying the number of hard drives to get extra performance by 30%, 40%. As we start to look at flash, we could shrink that number. So that would further dampen -- dampen the growth of total bits. But given that there's like 1,000 exabytes to play for, we have a long ways to go before we're ever worried about that part of it. But you have to recognize like disk as a technology is coming to the end of its life. It is not improving the way it used to. And people have been twisting themselves and knots trying to use it better for a decade now. And that's why this transition is going to happen and why it's going to flip.

One quick last one, Paul. The TLC, the hot layer versus the QLC, is the TLC SSDs, does that have higher performance than what you guys are offering in the QLC?

Bill?

Well, so TLC offers lower latency access than QLC. But one of the things that we're seeing is you can address the very high-performance workloads with QLC. So our AI offerings, we talk about EXA and FlashBlade. FlashBlade is one of the highest performance storage arrays in the world. It's all QLC. So we are confident that we can address a large portion of the market, the QLC. But in some ways, it doesn't matter. We have expertise in TLC. We've been shipping it on our DFM since 2017. If that's what it takes to hit that, we'll do that.

Thanks, Jason. Good multiparter. Asiya?

Sure. If I can just dig into the neoclouds opportunity, doesn't really get a lot of attention. I mean, I heard a lot about your scale AI. Just help us understand, I mean, there's some competitors there who are obviously dominating a little bit in that space, at least in all the media, and I've heard their presentation. So how does Pure play in that space? And can you maybe identify that opportunity and how big that could be for you guys as you kind of scale this out?

You bet. It's an interesting new market. In the past and also in my earnings calls, I've tried to put it in perspective in terms of overall size. In a $50 billion storage market, not even counting hyperscalers, it's still just a few billion dollars, right? But it's an interesting new area. Now Gen AI evolved from the HPC space, right, and the high-performance computing space. The high-performance computing space has always been an interesting niche in a number of different businesses, everything from the compute area, you remember -- you perhaps remember Cray Computing, right? And that was -- became part of HP. But also in terms of storage, it was a niche space as well that a couple of private storage companies played in.

One of the things about HPC was it was historically sort of like Hotel California, an organization would go in, could never extract themselves because the values in the HPC space, which were largely focused on scientists were very different than what enterprise values were. And you would get caught up in the benchmark competition that was taking place in that environment, which really didn't merge very well or suit itself to an enterprise market.

So we chose early on to stay focused on the enterprise market. Gen AI takes off all the practitioners come out of the HPC market. We were behind. I'd be very direct. We were behind in that space. Now we have always, I think, exhibited great discipline in making sure that we did not build bespoke software for every new segment that came along. So we looked at what we had, what capabilities that we had in the company, and we were able to then rally around our metadata engine and using that as a way of entering the market. We introduced FlashBlade//EXA earlier this year. We believe -- well, FlashBlade//EXA is now benchmarked as the fastest AI storage engine in the world by several factors. But we still have work to do in order to really get to the head of the pack from features and so forth.

So Rob, maybe you want to take it from here.

Yes, absolutely. Just to add a few comments to what Charlie said, just reinforcing the point, we've intentionally through the company's history, not started with a focus on HPC, but rather, as I mentioned in my presentation, the needs of the broader enterprise. What we have seen through now multiple new markets is the opportunity to leverage those investments, the IP we've built there in different ways to go address the needs of different markets. Said more simply, if you look at the HPC space, companies that start there have a lot of trouble moving into the broader enterprise. We've been able to go the other way and say, "Hey, let's start in the enterprise and now take some of those fundamental IP blocks and go address the needs of the neoclouds in this case.

One of the things that really was behind -- kind of the driving principle of how we built FlashBlade//EXA from the IP based on FlashBlade is we step back and we looked at what are the core strengths of FlashBlade? It's the software. It's the scalability that it delivers in the metadata. For the enterprise, right, it's like any design decision, you design for a certain space of possibilities. We deliver really, really good performance, really, really good economics, really, really good efficiency that covers the needs of 99.5% of the needs out there. What we have now done is to say, hey, if we go and take the software that's in FlashBlade, that scalable metadata engine, pair it with an open hardware architecture, we now have a very compelling offer to now supercharge the performance there to go after that last 0.5 percentile. And so that's really the evolution that we've gone through in terms of addressing the needs of the neoclouds. And as Charlie said, it's -- we just went generally available with that solution in Q2 and in early discussions with a lot of the top names and really good initial signs of interest.

Do you have a follow-up, Asiya?

Yes. I mean how should we think about the ramp of that, like as it goes through? I know you've put a lot in R&D and investments this year. And typically, when you had a strong year of R&D investments, that's kind of led to strong margin improvements the following year. So just give us a sense of this investment that you're putting in this year, how we should think about the ramp of that in terms of profitability into next year?

Well, let's see, how best to phrase this. We are in conversations with up to roughly a dozen neoclouds in terms of EXA already. I expect that we'll win some of those and increasing -- I think we'll increase our win rate over the course of the next 12 months.

I might also step back and just look at the broader context of our R&D investments and investments as a whole, right? I spent a good part of my discussion really unpacking where we add differentiated value across the stack and how we go and package that for the needs of different markets. So if we step back and look at where are we deploying R&D, where are we deploying engineers, it's into layers like Fusion. It's into layers like DirectFlash. It's into layers like our Purity software. And yes, a lot of those benefits flow directly through to the scale AI players in neoclouds. A lot of those benefits are driving the adoption of enterprise data cloud, as we discussed with Howard. A lot of those benefits are directly transposable into the needs of the hyperscalers. So it's a little bit hard to draw a direct line from R&D investments to the growth of a specific product. As Charlie mentioned before, we've been very, very intentional about how we direct R&D investments to get maximum leverage out of them to address the needs of the multiple different market segments that we're addressing.

I think we look at the enterprise data cloud as a gigantic opportunity for all of our established business. And we have several years of investment we're going to put into that new features, new capabilities coming out, evolving it and making it better year after year. So really, the biggest focus for R&D for us, hyperscale and enterprise data cloud. And then AI is -- and EXA really because AI is broader than just EXA. But EXA is probably the third.

Right. And I'll mention one other thing that may not be so obvious, but our focus on supporting the larger, if you will, breadth of enterprise and enterprise features is, in fact, starting to affect the way that purchasers specifically for AI like the GPU clouds are thinking about things. And what I mean by that is, remember, one of the things that we're well known for is Evergreen, the ability for customers to be able to upgrade, downgrade relatively non-disruptively to their environment without data disruption. Well, we're able to do that -- or we're promising that part of EXA will be that we'll be able to do that between FlashBlade -- regular FlashBlade, FlashBlade//S and EXA. And this is allowing us to have a much broader range of AI and consistent range of AI capability for smaller clouds, larger clouds, for clouds that are growing are not quite certain of what their needs are going to be tomorrow. And it's boosted interest in our overall product line, not just in EXA.

Yes. Just I'll comment on that. I was recently speaking with one of these research labs. And if you think about how in university research labs, how things get funded, their grant money tied to projects, like there's funds allocation to specific projects. And then you buy it, you sweat it and 5 years later, it's old, right? And then it's about raising grant funds. So this concept that Charlie mentioned of bringing Evergreen into that environment, when they get a taste of that way, they're like, okay, I sweat it and it gets better and it gets current over time as part of that same grant proposal and funding is something they've never seen before. We've never really like -- no one in the supercomputing market has approached that value proposition to that thing. And I'm starting to see that with the grant kind of funding that goes into some of these AI labs.

So let's go to Erik and then Wamsi and then Tim.

Erik Woodring from Morgan Stanley. I wanted to readdress the hyperscale question. So Kaz, Bill, Charlie. From my perspective, I guess, when we think about price and performance, performance is very clear with DFMs and Pure Storage. When I look at the HDD market, if performance was the factor that could disrupt HDDs, so to speak, maybe you would see something like greater adoption of dual actuators and HDDs. And you're not necessarily seeing that. So it still seems like it comes down to price. And I would love if you could maybe just elaborate on how you are increasingly able to compete on price as we look at this opportunity in the warmer and kind of colder layers of the hyperscale tiers because clearly, if you can bring price with performance, that seems like a no-brainer to me.

Well, let me start, Bill, and then...

Okay. I'll let you go at it.

I'd like to change the conversation because price, what price are we talking about, right? If you're talking about the raw price of the raw bit in the hard disk and then the raw price of the raw bit on flash, well, there's no competition. And that's consistently what a lot of the hard disk manufacturers will keep on arguing. It's not that price. It's the total cost of ownership of the system after you've put the whole system together. And when the total cost of the system for a hard drive environment is dominated by everything around the hard disk and not the hard disk itself, okay, we're not actually competing with the hard disk. We're competing with all the equipment that goes around the hard disk.

But Bill, let you.

Yes. So if you pay attention to the AI market right now, you'll notice when they talk about data centers, they're not so much talking about how many GPUs go in there. They talk about how many watts. How many gigawatt -- I'm going to do a 10 gigawatt factory or a 10-megawatt this or something. The reason that the conversation has changed is power is highly constrained, okay? So the cost of power is very high, right, because it's so constrained across the country and the world. but also the opportunity cost if you overrun the envelope, right? So if your data center uses more than it's available to it, now you have an exponentially high cost that is very debilitating for these people who are setting up data centers. So the power part of the equation or the cost equation is huge, right? And our solution is so much more power efficient. So think about fitting within that power envelope and then lowering the amount you spend on cooling and then shifting that power resource to something else is very valuable to our customers today.

And then -- so the flash modules are roughly 10x the size of a hard drive now. Okay. The $20,000 server or $30,000 server in which you house some hard drives, you need 1/10 as many of those. the top of rack switches that you use, you need 1/10 as many of those. The power supplies, the power distribution units, you need 1/10 as many of those. The amount of human effort that it needs to replace broken hard drives and the disruption, you need less of those because the hard drives break more often, you actually have extra inefficiencies in the erasure coding that you use to make the data resilient. So you gain efficiency there. And all those efficiencies add up to where the cost and the flexibility is much better and that's with what we're shipping today. Then you go forward and you say, okay, so today, we're 10x the size. Next year, we'll be 15x the size. That's assuming they grow 33% and the people could use the extra space.

Year after, we'll get to 20x or 25x in size. And just completely wipe out that cost difference. And then again, the power budget, if you overflow the power budget, the cost is gigantic. If you think about an enterprise data center, they all have power budgets coming into their data center. And when you have to contact Con Ed or PG&E or whoever your utility is and say, I need another transformer and I need you to like dig up the street and upgrade the power like forget it, your costs are millions. And there's nothing. Storage is, generally speaking, the #2 consumer of power in data centers. And we're taking that 25%, 30% of power in the data center and taking it down 80%, 90% or more, that's a gigantic advantage when everything is budgeted and once.

Erik, let me just put it in perspective for you because we have to think about the hyperscaler industry as a very capital-intensive industry, right? You know that most of them spend about $7 billion to $10 billion per year on CapEx, okay? All of that spend for the vast majority of that spend goes into data centers. So when you are someone who is managing that level of investment, you want that investment to last for a very long period of time. And all those costs that Kaz spoke about create a limit to how long your data center investment is going to last for. And we're pushing that limit out because we are a lot more efficient in the technology we provide that addresses the power constraints that data centers face today. So we have to think about it in those terms from a capital intensity standpoint, we better the capital intensity of hyperscalers.

And our technology has a much longer lifetime. We've been shipping DirectFlash modules for over 9 years now. And the ones we shipped 9 years ago are still like amazingly reliable compared to hard drives. Whenever we do talk to people about our annual failure rates on it, we're including those as well as the brand-new ones and what we're seeing. And that extra lifetime gives them a lot of options that most people aren't even yet taking advantage of. That will further extend the gap.

Okay. And so that is super, super helpful. I'd love to maybe just put price and performance then to the side and say, where do the bottlenecks exist maybe when we talk about go-to-market and disrupting all of this, right? This has been an established market for years. You guys are the new entrant, you're the disruptor. How does the disruptor -- again, when you put those factors to the side, not to say they're not important, but how do you make your entrants? What is the selling path into these hyperscalers? And how long does that take? I'm not asking about any specific customers, but just like what is the bottleneck maybe if we get?

So thank you, guys. First off, I want to -- I'll start with -- we've been doing this for years. It feels like we started last month, but we've been doing this for years. And we set up a separate line of business, hyperscale line of business 2 years ago. So completely separated it because we had had a bunch of success, and we're well into this. The hyperscale -- getting into hyperscale involves showing a value and improvement to the engineering community. That's really how they think about things. You have to go to their engineers and their engineering leaders and get them bought into the design and the approach.

As far as disruption, most of the hyperscalers are very into disruption. They want to see a new technology that transforms and gives them a competitive advantage. So they're very open to those talks of like, hey, what if we changed everything. They do like that. Nice thing is our technology doesn't require them to change everything, right? It's easy to slot in. We have an interface that they're used to seeing. And if they want it to be a little different, we can change it for them. So we can disrupt the cost and the performance and the reliability without being hard to ingest.

I will say this is the -- that was the most optimistic I've heard Bill talk about this space. There -- we do battle as we would in any organization, any enterprise organization and Not Invented Here syndrome. And -- or a -- well, we've always done it this way, and this is going to change a lot of things. And so we have that in every -- almost -- I think almost every situation. And then, of course, they have their own pace at which that they are doing their own -- when they design a next-generation data center, it's a product for them. It's like Apple developing another -- the next-generation iPhone. Everything is engineered for iPhone 16, iPhone 17, right? Everything from the camera to the software, to the marketing and everything else. And that's how the hyperscalers design their data centers. So we also have to get on their cadence, right? And you either get at the front end of their cadence and therefore, it can be part of that design process or you've missed it and then you have to wait for the next time that comes around them.

Remember, a hyperscaler never changes their data center. So they plan a data center, they design every bit of equipment that's going to go into it, every watt, every bit of cooling, they design it, they build it, it stays the same and 5, 6, whatever years later, they got the data center. They might keep the shell of the building, but every system in it changes, all the compute, but also the cooling, everything. And that's very different. In enterprise, they have a data center and they're rolling in new equipment all the time. And so it really is intercepting that process of designing the data center, which, as Charlie said, it's like a product. So that does also take time.

Let's go to Wamsi.

Wamsi Mohan, Bank of America. A few questions. I guess, one is just on -- following up on the go-to-market question. If we think about the product disruption. I mean, I think Pure has shown product disruption now for many years. But when we think about market share, the incumbency that some of the legacy players have in the storage market has been hard to overcome in some ways. And so do you think that now is a time where this re-architecting of the data center has reached a tipping point where you're able to go and really make a change and drive incremental share? Do you think that your go-to-market actually has to change based on the product portfolio improvements? How do you go about capturing incremental share today?

Briefly, I think -- and then I'll let some of the other executives speak. I think yes and yes. One is I think that AI has been a boon in a sense because I think it's forcing customers to rethink their overall architecture at a higher level. I mean the one thing about storage, it was pushed down to the storage admin on decisions. Nobody -- no one in the higher levels of IT really wanted to discuss it. Part of it was because it was so specialized, right? But now they have to rethink their overall data architecture, and that's a senior level strategy discussion and planning. And those -- so those conversations are opening up this opportunity. So I think we're fortunate. We were going down the path of Fusion and Enterprise Data Cloud many years ago because it's a better way of enterprises managing their environment, but AI has come along at a point in time that's also getting them to consider these things at a higher level. So I think that's been very positive for us.

At the same time, yes, the go-to-market now is very different. again, I'll go back to storage being bought and sold the same way for decades now. And so your best sellers are used to speaking about it in a certain way. They actually know for a fact that that's the best way to sell storage. And what we're saying is, no, that's no longer the best way. We now have to speak the language of data, data management and all of that. So the training, the retraining, the focus on more of a franchise sale rather than an individual use case sale, this is all very different. And it's a lot of investment. This -- we haven't -- we talked about investment in technology. There's a lot of investment in the go-to-market that's going on.

Yes. No, I want to elaborate on what Charlie has said because this is a beautiful question. The world is changing because AI provides the impetus for the change to standardize storage, right? And therefore, in the way we go to market and the way we sell, we have to have a conversation that enables that standardization over the Pure platform. which means that if you are going to sell to a large organization, I'm picking your bank, for example, then the conversation has to also do with the legacy environment that exists, right? So part of bringing together franchise deals is helping customers migrate away from the legacy environment and dealing with potential write-offs that may ensue should they standardize over the Pure platform. So it's a very different sale. You no longer sell down deep into the organization to the IT department. You have to involve the CFO because there's a lot of financial engineering required. There's a lot of financial implications attached to the sale. So we are on that journey.

Yes. And I can double-click how we project it changing over time, right? Because as you think about this platform effect and you think about what SaaS companies have gone through, yes, there's the landing the change disruption cost. But given our Evergreen and our long-lived hardware. People have like asset depreciation cycles, 5 years, et cetera. For some of our existing customers, we've already seen them adopt to longer asset value depreciation cycles. When you get to 7 or 10 years, it fundamentally changes the unit economics and cash value and asset depreciation. And then you think about asset utilization in SaaS companies, like when you think about SaaS companies, you end up with customer success driving expansion utilization. So the go-to-market, as you start thinking about platform effect will evolve into more of this land, expand customer success motion, et cetera, and the capital asset efficiency you get on some of our long-lived components with Evergreen is tremendous. Now you have to get them into that world, but it creates this nice platform effect.

I have a follow-up on their SaaS company comment. A lot of software names have announced zero copy data sharing. Is that something that you think impacts the footprint that you have at these companies at all?

It's a great question. We announced something similar to that. Well, 2 things. One is, as I mentioned, Data Cloud, we can allow customers' production data now to be available as their data warehouse rather than copying the data. We also have zero move tiering, which is a similar kind of concept. Customers can have a balance of performance and low cost for storing information and leverage that across the board. I guess our point of view is however we're able to help our customers economically to make better use and more efficient use of their data, it's going to inure to our benefit. So by the way, we've been doing this for years. Every time we come out with a new major software release, we generally improve upon our data reduction, our data compression. So effectively, what we've done is we've taken a customer that might have bought 100 terabytes 5 years ago, and now it's storing 200 terabytes. We didn't sell them another 100 terabytes. We gave them free software upgrades that doubled their overall capacity. We're happy to do that because it gets us great scores and allows us -- we grow nonetheless.

Wamsi, would you please pass the microphone over there to Tim?

Tim Long at Barclays. Two as well, if I could. First, I wanted to touch on Evergreen//One and the SaaS offerings kind of relative to hardware. It's been a good transition, but I'd say it's probably been a little unpredictable. So can you just talk about kind of the transition in the overall model you're talking about today and as AI gets to the edge, does that change -- I'm guessing the decision now is CapEx versus OpEx, not necessarily technology. So when we move forward into this new era, do you think that SaaS versus CapEx sale changes? And then separately, I did want to just touch on, I think Pure traditionally does very well with kind of higher-end enterprise SaaS companies and banks and things like that. Just curious how those companies are looking now because some of those SaaS companies could start to look like neoclouds or much larger in the future. So just curious what path those companies may be on.

Let me start and Prakash, maybe pick it up. It's a very dynamic market for sure, in many ways. And you're right. I mean our ability in the past to be able to predict what percentage of our sales would go down the route of Evergreen//One versus traditional product purchase has not been -- we've not been able to be highly accurate. A lot of it -- part of the issue there is it's driven a lot by macroeconomics. It's driven by everything from interest rates to uncertainty in the market, uncertainty of growth. The more of the uncertainty, the more they like Evergreen//One, the more certainty or at least the more the sense is that they have certainty, the more they like a standard product sale.

We're still in the early phases, I think, of enterprises appreciating a SaaS model in a private data center environment. And so that understanding, when I say that understanding, I mean, up through the financial organization inside the enterprises is still very immature. And so we're also seeing the immaturity, if you will, of a market that tends to make prediction or forecasting a little bit more challenging. So I would say all those things have played into it. Now it does continue to grow, but the specific growth rate quarter in, quarter out, even annually is buffeted by all those things that make it difficult for us, even for us to predict.

Yes. Like if I kind of break it out into the 2 parts of your question, right, there -- we've had broad adoption across a wide variety of verticals, but some people have been adopting for the financial value, other people for the operational value. right? If you tie it to the financial value, it's tied to all the -- like our -- as economic uncertainty increases, the offering is much more compelling on the financial side of it, right? Now I do see this enterprise data cloud approach and this entrance of AI starting to elevate the operational value more and this AI insertion point is creating some clarity on this operational insertion point. The place where the operational insertion point was most pronounced is where you mentioned.

We got 6 verticals from high-tech SaaS to some of the more forward-leaning technology guys that were coming from the standardization. If you're coming from a SaaS world, you're like standardization at scale is everything, right? So we didn't need to sell them on that. That is the way they operate, and we were fit for purpose into their operating environment. They get the idea of reducing variability. And there has been -- if you couple that with messages around, hey, cloud -- going to cloud, cloud repatriation. Those things you've seen that kind of in the press, et cetera. But as you get to certain economies of scale, we've also seen a driver of some of that standardization and high-tech SaaS being repatriation workloads as well.

I might add one more thing to that, Tim. We talked about go-to-market before. There's another element of introducing the full as-a-service model that we need to consider, which is it's a new motion for people. It's a new motion for buyers. It's a new motion for sellers. It's a new motion for partners. And we've seen really nice progression over the last several years as we focused on educating our sellers, really all of the parties at stake to the benefits of this model. Put more simply, several years ago, I would go out on a sales call and our salespeople, very well meeting might present an offer to a customer and say, "Hey, this is the traditional product sale or if you want, you can buy it OpEx. It looks like this.

And when you -- to Prakash's point, when you present the offer in that manner, you're missing out on a ton of the value, and so somewhat surprising to me anyway is that although the enterprise is well used to consuming in a service-oriented model in the cloud, it's still a relatively newer motion when that equipment sits on their own floor tiles. And so we've been making steady progress, both with our sellers, our partners. And I think now to Prakash's point, some of the technology kind of market factors are illuminating the buyers to, hey, there are additional benefits here with the full as-a-service model.

So we're getting close to the top of the hour. So Nehal, you'll have the last question.

Yes. Nehal from Northland Capital Markets. Charlie, in your discussion of evolution of Pure Storage's entry into the high-performance market. You talked about that there are still things that you want to do to get to the head of the, what I'll call the AI native data management market players. Can you talk about what are some of these features that you perceive that you guys need to bring to the market still?

Yes, I'll speak about some of them. They're all quite straightforward. We provide one type of, for example, high ability, and there's a market for 2 or 3 different types of high availability. So that's -- we're going to be developing more and more of that capability that we're putting out this -- by the end of this year is multi-tenancy. That's an important part. And then like in any market, there's a whole -- there are sets of features that different customers need for different purposes. And my point of view on this is that we probably have about a year's work to get us to the head of blind on those different -- the capabilities that we're currently missing.

Is being able to access the metadata storage pools that are not owned by Pure Storage part of that road map?

Not for this specific -- not for the specific use case you're talking about, which is the AI data pools, okay, for the large-scale AI environments. I think over time -- right now, our -- what we're calling our enterprise data cloud is very much focused on Pure Storage, right, Pure's products. I do think over time, and customers will want the ability to be able to also pull into that management framework storage that sits in other areas, not just maybe competitive products but areas like SaaS as well. And that's certainly an area we're going to be focused on over the next couple of years.

And on that, we do have -- we do reach in, in container workloads to storage sitting outside through our Portworx. So in that heterogeneous storage space across both cloud and on-premises, we see that demand in the container space to be higher, and we're leaning into Portworx as part of that.

Thank you very much. Before we break, Charlie, did you want to do a sum up?

Yes, I just want to thank you all again, hopefully -- and of course, there will -- some of us are going to have to run right away because we have the larger customer event, but I think some of us will be able to stick around for a little bit of informal Q&A. But I want to thank you all for your time. I want to thank everybody that has been online for your time. I hope we -- I mean, there's a lot going on, and I know there's been some confusion. I hope we've clarified a lot of this. Certainly, we'll be referring to this in our -- in future meetings and earnings calls. But I want to thank you again for your time. Take care.