PSTG - NYSE

Good day, and welcome to the Pure Storage First Quarter Fiscal Year 2024 Earnings Conference Call. Today's conference is being recorded. [Operator Instructions] At this time, I'd like to turn the call over to Paul

Thank you. Good afternoon, everyone, and welcome to Pure's first quarter fiscal 2024 earnings conference call. On the call, we have Charlie Giancarlo, Chief Executive Officer; Kevan Krysler, Chief Financial Officer; and Rob Lee, Chief Technology Officer. Following Charlie's and Kevan'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 dynamic 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 relating to our business is contained in our file 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 second quarter fiscal '24 quiet period begins at the close of business, Friday, July 21, 2023. With that, I'll turn it over to Charlie.

Good afternoon, everyone, and welcome to our Q1 FY '24 earnings call. Thank you for joining us today. We were pleased with our Q1 performance in what continues overall to be a challenging IT environment. Highlights for the quarter include the highest all-time sales for Evergreen//One, our first sales of FlashBlade//E, which is experiencing the fastest first quarter pipeline growth for any new pure product. Our largest single order since inception of almost 8 digits for our Cloud Block store product and the release of a major upgrade to our FlashArray unified block and file software. We are especially pleased with the customer response to our E product line. The E line is the first and only all-flash storage system that can address the secondary storage market, at competitive prices to 7,200 RPM hard disk systems, but with only 1/10 the power, space, cooling and labor requirements. As I've stated in the past, the days of hard disks are coming to an end. We predict that there will be no new hard disks sold in 5 years. But beyond the benefits of the E product line itself, it enables Pure to now compete for our customers' entire storage environment. It enables Pure for the first time to be our customers' complete storage partner. Something that our customers have been asking for, for years. The operational and economic benefits of Pure's comprehensive storage portfolio are clear and overwhelming and are based on sustainable technology and business model advantages. In March, we gathered our sales team for our annual sales kickoff, an event we had not held in person since 2020. Everyone was energized by both the event and the training we conducted there. This ongoing training focuses on honing sales skills with our expansion into the secondary storage market, advancing our as-a-service offerings and selling in today's environment of constrained IT spending. This valuable training is already bearing fruit and enabling us to reach new customers, better support the needs of current customers and reduced sales cycles, which have lengthened in this economic environment. It's clear that our continued innovation strongly resonates with our customers, whether it is for AI/machine learning, rapid recovery from ransomware, high-performance databases, electronic design automation and video editing traditional or cloud-native applications and now also for secondary storage environments such as content and media stores, enterprise imaging and even traditional backup and archive. We deliver unique outcomes that are highly valuable to our customers in every environment. Compared to our all-flash competitors, we are 10x more reliable. We are 2X to 5X more power and space efficient, and we require 5 to 10x less manual labor to operate, resulting overall in at least 50% lower total cost of ownership. Also, our products never become obsolete and never require forklift upgrades because our Evergreen program provides continual hardware and software upgrades nondisruptively, to the customer's application environment forever. Products we sold 10 years ago are not only still in service, but have been continually modernized to our latest models without disruption or additional customer expense with our Evergreen subscription. These capabilities are based upon 4 unique and sustainable competitive advantages. One, our purity software uniquely works directly with raw flash, while other competitors use more expensive, less efficient and shorter-lived SSDs; two, Pure's highly consolidated product line consisting of our common operating system, Purity and 1 management system Pure1, operating on both a scale-up and the scale-out platform utilizing common direct flash modules, while competitors require many disparate software and hardware platforms to cover the same breadth of use cases; three, Pure's unique Evergreen technology and services, which guarantee that deployed products never become obsolete, never need to be replaced and enable nondisruptive upgrades. And four, Pure's cloud operating model, which enables customers to operate their storage the way that cloud customers operate theirs, highly automated, orchestrated and available as-a-service. As we expected, Evergreen//One is thriving in this economic environment. Sales of Evergreen//One more than year-over-year. As a reminder, Evergreen//One is Pure's Storage as a Service offering that enables customers to access storage entirely through service level agreements with no capital expenditure only paying for capacity as they use it. Customers can place their data on-premise, on Pure-owned infrastructure or on AWS or Azure with Pure's Cloud Block Store. The customer only pays for what they use under a single contract for enterprise-class capabilities for less than what they pay for raw cloud storage. This past quarter, we saw the largest individual sale of Pure Cloud Block Store at almost 8 figures, a Fortune 500 health care organization purchased Cloud Block Store because of its ability to securely store data in the cloud with enterprise features, reduced management overhead and lower TCO. By using Cloud Block Store, the organization is able to significantly reduce their cloud storage spend while getting the most out of their data. Our extraordinary lead in driving power space, labor and e-waste reduction both on-prem and in the cloud has also garnered attention amid increased customer focus on the selection criteria. The continued strength of FlashArray//C and interest in FlashBlade//E speaks to our customers' demand for the TCO benefits of Pure all-flash products over competitive offerings now including both their flash and hard disk systems. In particular, FlashBlade//E consumes approximately 1/10 as much power and space as similar capacity hard disk systems that it replaces, requiring up to 1/10 the labor and generates less than 1/10 the e-waste. Only Pure's direct flash management and operational simplicity is able to deliver this operational performance. As I mentioned, Early interest in FlashBlade//E is off the charts for a new product. FlashBlade//E is the second in a series of products that can compete for the secondary tier and some lower tiers of the storage market entirely dominated today by hard disks. Prior to FlashArray//C and FlashBlade//E, all flash products were only price competitive for high-performance systems and therefore, Pure could only provide products for our customers' Tier 1 storage needs. With the introduction of our E product line, Pure can now compete for customers' entire storage estate, enabling Pure to become their complete storage partner for the first time. For years, customers have asked us for products that could address the remainder of their storage and space. I have now had many customer visits since our introduction of FlashBlade//E with senior IT executives describing our key advantages and our ability to provide flash solutions to their entire storage environment. A common question from these senior executives is why aren't we doing this already. In April, we announced a major update to our FlashArray unified block and file software, representing a significant expansion of our broader file strategy and portfolio. I'm proud to share that we're now able to address customers' file needs across high-performance, general-purpose NAS, VMware over NFS and dozens of other use cases, allowing us to compete for all of a customer's file storage as 8. Best of all, FlashArray customers can simply upgrade to the latest software to get these capabilities without any additional expense. This unified offering was a key component in the largest individual international market win in Pure history last quarter. Touching upon the most recent trends, generative AI and chatGPT has brought artificial intelligence to the top of mind in all of our major customers and has become a focal part of literally every earnings script this quarter. Pure saw the AI opportunity years ago and started innovating in this area with our introduction of FlashBlade in 2017 and then with our AI-ready infrastructure Aerie product co-developed with NVIDIA. We've continued to advance FlashBlade's high-performance parallel architecture, and Pure continues to be the go-to partner for storage on AI projects. For instance, we support more than 10 leading autonomous vehicle development companies in managing and processing the massive amounts of data required for their machine learning activity. In addition to our very successful position with Meta in their AI research supercluster or RSC, the largest AI supercomputer in the world. Pure is the chosen vendor for AI environments across a broad range of industries, including media and entertainment, pharma, health care, aerospace, transportation and financial services. We expect our leading role in AI to continue to expand, but we are equally excited that the requirements for Big Data will drive even more use of high-performance flash for traditional bulk data. As we mentioned last quarter, we expect the current macro environment to continue through this fiscal year, and we continue to operate the company with our usual diligence improving productivity and focusing investment on meaningful innovation and growth. Our as-a-service offerings, including Evergreen//One and Evergreen//Flex and our Pure financing vehicles provide customers with a wide range of economic alternatives to address their business needs. Pure's superior TCO and flexible Evergreen offerings are making a difference in this challenging IT economy. While we saw continuing caution by enterprise and cloud customers in Q1, similar to what we saw in Q4. We also experienced enhanced demand for our most cost-effective studies, especially Evergreen//One. Given all of our advantages, we remain confident that we will continue to increase our market share, outgrow our competitors and pick up even greater momentum, especially as our new products and services gain mind share. I am confident that we are gaining recognition with both customers and prospects that Pure is the company to trust for their future data storage architectures. We are years ahead of the competition in our ability to provide all storage needs with the most consistent, modern and efficient storage solutions. We enjoy a highly sustainable competitive advantage based on the only direct to flash operating system in Purity, a simple, consistent product line with common management, our Evergreen technology to continually upgrade our products nondisruptively to the third state of the art and our ability to provide our customers with a cloud operating model. our new capability to compete for the full range of enterprise storage needs gives us even greater relevance to our enterprise accounts and enables us to deliver a full and far more integrated storage solution to our customers. In closing, I am excited to share that in just a couple of weeks, we'll be hosting our Annual Accelerate User conference in Las Vegas. We're looking forward to seeing customers, partners and analysts from around the world to discuss the future of data storage and management. I'll now turn the call over to Kevan.

Thank you, Charlie, and good afternoon, everyone. In Q1, we achieved revenue of $589 million and operating profit of nearly $20 million, exceeding our expectations. We also set an all-time record of Evergreen//One subscription sales this quarter as demand was exceptional. We were pleased that our U.S. enterprise business exceeded our expectations this quarter. Macro conditions continue to be challenging, consistent with what we saw in Q4. Against this macro backdrop, our sales force and leadership are actively monitoring deals to get ahead of challenges as well as continuing to focus conversations both on our business value and total cost of ownership advantages which are unmatched against our competitors. Our subscription services annual recurring revenue grew 29% year-over-year to $1.2 billion and subscription services revenue of $280 million represented 48% of total revenue. Remaining performance obligations, or RPO, grew 26% year-over-year to $1.8 billion. Similar to the remarks we've made in previous quarters, our RPO included an outstanding commitment with one of our global system integrators. During Q1, this remaining outstanding commitment was fully satisfied with Evergreen//One sales. When excluding the impact of the past outstanding commitment from our global system integrator, RPO grew 31%. Our headcount increased slightly to approximately 5,270 employees in Q1 and and we remain disciplined in managing our costs, including hiring. Incremental investments in headcount remains focused on quota-carrying sales capacity and critical business hires. As I previously mentioned, total revenue in Q1 was $589 million and product revenue was $309 million. As we noted in previous earnings calls, Q1 revenue last year included $60 million of product revenue that was contemplated in the second half of last year. Excluding this impact, Q1 total revenue grew approximately 5%. U.S. revenue for Q1 was $427 million, and international revenue was $162 million. We also acquired 276 new customers during the quarter. We were pleased with our continued strong gross margin performance in Q1 of 72.2% with product gross margins of 70.8% and subscription services gross margin of 73.7%. Q1 operating profit of nearly $20 million exceeded expectations and included higher year-over-year costs for salaries and our first sales kickoff event since 2020. Pure's balance sheet and liquidity remains very strong, including $1.2 billion in cash and investments. In April, we reduced our overall debt paying off $575 million in convertible notes using $475 million in cash and $100 million from our revolving line of credit. Cash flow from operations during the quarter was $173 million and capital expenditures totaled $51 million. In Q1, we repurchased 2.9 million shares of stock, returning nearly $70 million to our shareholders and have approximately $211 million remaining on our existing $250 million repurchase authorization. Now turning to guidance. We are reiterating our annual guidance for FY '24 with revenue growth in the mid- to high single digits and expect an operating margin of 15%. Our annual revenue guidance assumes that macro conditions will continue to be challenging and will be consistent with what we have seen over the last couple of quarters. We expect continued momentum of our Evergreen subscription services in particular, Evergreen//One. The strength of our Evergreen//One offering has been contemplated in our annual revenue guide as the recurring revenue for these services is recognized over time. Also, as Charlie mentioned, early customer response to FlashBlade//E, which became generally available in late April, has exceeded our expectations. Our FY '24 annual revenue guidance that we provided last quarter assumed a modest revenue ramp during the second half of the year from sales of FlashBlade//E. While we are very pleased with the early response to FlashBlade//E, our FY '24 revenue guidance continues to assume a modest revenue ramp during the second half of the year. Moving to Q2 guidance. We expect Q2 revenue of $680 million representing an increase of approximately 5% year-over-year. Our Q2 revenue guidance implies continued strong subscription revenue growth and a slight year-over-year decline in product revenue. We also expect Q2 operating profit of $90 million as we remain focused on profitable growth and ensuring we are appropriately aligning our cost structure with demand. In closing, through our innovation, our competitive advantages are clear and aligned to our customers' focus on both performance and cost. We are uniquely positioned to deliver significant business value while reducing our customers' total cost of ownership, including labor, energy and real estate. It's a pleasure to also invite you to join us for our product and technology-focused Financial Analyst Meeting at Accelerate on June 15, either in person at Las Vegas or virtually through our Investor Relations website. With that, I will turn it back to Paul for Q&A.

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

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

Congrats on a really strong print here despite the tough macro environment. Charlie, I was hoping you could talk a little bit more about what are you seeing from an AI infrastructure investment perspective from your customers? You folks obviously have good engagement with Meta that's been doing well, I think. But I would love to hear how your customers really thinking about their storage needs broadly and very specifically around Pure's product portfolio around IRS help to put any dimensions around it? And how do you distinguish yourself from the requirements?

Yes. Thank you, Amit. Good to hear from you. Well, we view AI -- well, we've been doing AI as a great opportunity for us and representing a reasonable portion of our sales now for multiple years, especially with the introduction of our FlashBlade product 5 years ago. And we've continued to expand that area. And we do believe that we are the go-to partner as I mentioned for AI projects, and we literally support well over a half dozen to a dozen different industries in their AI activities. The majority of the AI now continues to be part of what we -- as an industry and what everyone on the call had known of AI for many years, whether that's genomic research or advanced analysis, financial analysis or self-driving cars and so forth. Now what's happening, of course, is every company is looking at large language models, chatGPT, et cetera, trying to determine exactly what it means for them. We've seen some interest in that area, but it still remains a minority. The majority being traditional, much -- if I can use that word, with AI traditional AI projects. But we're most excited by is both the opportunity for a high-performance FlashBlade systems. But frankly, we are at least as excited that customers now more and more are going to want to put their data that is in hold, hard-to-reach hard disk systems and make that available for analysis by putting it in much more higher performance flash-based systems. And both are FlashArray//C and FlashBlade//E are perfect repositories for that. So we think it's coming out at a perfect time. And let me just -- let me go back to Meta. We continue to have an excellent relationship around AI with Meta. They recently fully turned on the first 2 phases of their research supercluster, which they announced a few weeks -- a few weeks back, and we look forward to continuing to work with them on that project as they continue to build it out, but also on other projects that they are contemplating in the other parts of the company. Rob, anything to add?

No, yes. I think a couple of things. Number one, Amit, as Charlie mentioned, we're very excited about really what we see as two sets of opportunities that AI creates for us and I think are very constructive for Pure supporting the AI training environments themselves as well supporting enterprise customers as they look to connect their data sets to AI-powered applications. One set of demands, which Charlie discussed, is hey, this needs to go in store larger and larger amounts of data. It can't be cold data. And so it's a perfect fit for C and our E products, but then equally so, enterprises are going to need to connect data from all across the organizations and all across different silos of infrastructure into these applications. They can no longer be islands of their own relegated to silos that the infrastructure had held them to. And if you step back from it, those are the hallmarks of a cloud experience for storage and is exactly what we're delivering with the cloud operating model. So net-net, we think this is very constructive for us, both in supporting the high-performance and large-scale training environments certainly with our secondary tier disk takeout product lines, but also in what we're delivering with the cloud operating model.

Our next question comes from Meta Marshall of Morgan Stanley.

Maybe first question, you noted that you had had some FlashBlade//E sales have begun. Just wondering if there was any kind of surprise on where that uptake has been. And then maybe as a follow-up question, you noted kind of the 8-figure block store deal. Just wondering kind of how long that deal has been in the works and kind of what were the ultimate decision-making factors for that? Appreciate it.

You bet. Thanks, Meta. In terms of the -- let me start with the Cloud Block Store. As you know, we introduced that product about 2.5 years ago approximately, and continued to work with major customers in terms of their efforts to so called lift and shift their traditional apps into the cloud. . That progress of lift and shifting traditional apps out has probably taken longer than any of our customers expected, perhaps a bit longer than we expected as well. That being said, now that, they -- now that, that has started and they're seeing some of the bills coming back from the cloud vendors for the storage of that, those large amounts of data for traditional apps, now really start to become even more curious about Cloud Block Store. This particular deal, we had talked to them about 1.5 years ago before they moved into the cloud. And they -- at that time, they were so busy moving into the cloud, and they didn't really -- I don't think they really appreciated what it would cost them as they started to deploy a production environment. As they deployed that production environment and started to see the costs, they came back to us and said, "We really would like to understand better the Cloud Block Store. And I would say from that point where they came back to us to the sale was probably only about 3 months. So a relatively short period of time.

Our next question for today comes from Tim Long of Barclays.

Charlie, I was hoping you could talk a little bit about visibility. A lot of talk of macro. And obviously, last quarter, there was a little bit more challenging environment, but pretty positive that you guys are keeping the full year here. So can you just touch on kind of how your visibility compared to a few months ago? And what are the kind of factors that can kind of swing the numbers into the second half or end into next fiscal year, maybe from a product standpoint or where there could be upside?

You bet. What we saw -- if I go back to Q4, we saw a fairly sharp degradation midway through Q4. And what we saw in Q1 was, if you will, a stabilization to what we saw at the end of Q4. So it didn't get any better, but it didn't get any worse either. The visibility is basically just the way we're handicapping, if you will, how -- what we see in terms of -- from our sales team and the length of time that we believe accounts will close. And I think our sales teams have become much better at really understanding what the full gamut, if you will, of signatures and approvals that they're going to need to get in each account and therefore, have also become better at handicapping and staging, if you will, of the deals. So I would say that in general, the current environment, we see as stable. We would hope for improvement towards the year, but we're not counting on it at the moment. And I think that generally, we feel that this will be a -- and I'm going to exclude, let's say, surprises in federal issues. But if we see -- we're expecting stabilization through the end of the year and hopefully an improvement towards the end of the year beginning of next.

Tim, I probably would add a couple of things. This is Kevan as well. We did indicate the fact that our enterprise business performed better than expectations. And again, I think that's a testament to our field really adjusting to our customers' buying behavior. So that's a plus for us. I think -- the other two key highlights for us that came across as incremental strengths, if you will, would be Evergreen//One performance, which we alluded to. Again, that strength was much stronger than we were even anticipating and we were already anticipating in this environment. Our Evergreen//One storage as a service sales would be strong coming into this type of environment. And then FlashBlade//E, again, is a highlight for us. Early, but customer response has been fantastic. And Charlie, I don't know if you want to say a few words on FlashBlade//E in terms of what we're seeing there.

Yes. As we mentioned, it's a very fast pipeline build. I've been in front of dozens of customers now and the excitement with customers around this product line and with the prospect of replacing their disks, which are troublesome, with all flash products has been very high. And as I mentioned, I think what's most exciting is that we're seeing customers appreciate the fact that we can address the majority of their storage needs and to be able to do that with the simplicity, the power, the ease and the reliability of Pure products. So that's -- I have to say the enthusiasm that when any one of us go in with the new pitch, if you will, to our customers. that we feel coming out of it is really palpable.

Our next question comes from Pinjalim Bora of JPMorgan.

Great. And congrats on the strong quarter. One question for Kevan. Subscription, obviously, is very strong, and you kind of highlighted the potential headwind to revenue because of that. Is it possible to quantify that? You've kind of beaten the guidance by $29 million, you're kind of keeping the full year, which I appreciate, but I'm sure there is a little bit of conservatism there, too. I'm trying to understand what for the year, what would you kind of circle as a potential year-over-year growth headwind from that Evergreen//One strength?

Yes. Look, we won't get into specifics there, Pinjalim, consistent with our practice of really talking about the subscription portfolio in total as well as from a performance standpoint, we do the same from a product standpoint. But you are thinking about it right in terms of the fact that the Evergreen//One strength we'll see it come through on top line over time, which is a positive in terms of quantifying that. We won't -- we'll let you do the math specifically on that. But again, from a scale perspective as well, we already have over 1/3 of our revenues are subscription. So the headwind isn't as intense as one would expect though it does have a consideration for us, and it has been contemplated in our annual guide, which we reiterated.

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

Yes. I was wondering, just to clarify, are you embedding anything from Meta in your guide? And as my main question, I want to ask you. We're hearing a lot of strategic buys that are happening in -- now for some areas within memory, particularly in NAND. And wondering how you're thinking about it and potential impact to margins for the rest of fiscal '24? Or maybe a different way to think about the same thing as you are going to see competitors particularly be able to take advantage of this low-cost land environment for maybe a prolonged period of time? Do you think that closes any of the competitive gap at all or pressure to margins in the second half?

Yes. Maybe we can take -- talk about the competitive advantages. First, Charlie, if you want to take that around our direct to flash management advantages, which really, I think, are sustaining. And we'll hit that first, then we'll hit Meta.

Absolutely. Well, the interesting thing here in -- I'm sorry, Wamsi is that we believe that our advantage based on Purity operating system and our continuing advancement of our -- what we call our direct flash modules is going to allow us to accelerate our advantage over SSDs. So I'm going to ask Rob to jump in and give you some detail there. .

Yes, absolutely. Wamsi, I think we've been pretty clear with our view that disk is -- well, a dead technology spinning, so to speak. And -- but our view on SSDs, frankly isn't that much rosier. We think it's SSDs are going to fall further and further behind. And that's really driven by our ability to distance SSDs with our direct flash technology. If I step back, certainly, SSDs have played an important role in making flash more broadly available, but they're inherently less efficient, more complex, less performance, less reliable and just have overall shorter lifetimes than the systems we can deliver based on our direct flash software technology. And so while we get a lot of benefits today, the issue for SSDs and frankly, our competitive set that are reliant on SSDs is that SSDs are going to find it harder and harder to keep up with the rate of improvement that we're forging down. If we think about it, NAND flash, it's getting harder and harder to work with, that's creating pressure for what the SSD has to do from a technology perspective, trying to build larger capacity SSDs only exacerbates that. And then I think you have an economic barrier around, hey, is there consumer demand that's going to drive the same -- and follow the same type of road map that we're going to drive with our direct flash technology. And so you net it out, I think we're very bullish here that we've got a 3- to 5-year structural and sustained competitive advantage over, frankly, the rest of the field that I think is trapped on SSD technology. And we're going to be pretty aggressive about going after that. And then bringing this back to Meta, right? I think that our engagement in Meta is a great example and proof point of the value that we're able to deliver based on this technology, as we've discussed in prior calls, one of the -- really, the key reason that we won that engagement was our ability -- our solo ability to deliver the balance of performance, cost efficiency as well as power space and cooling savings. And that traces its roots directly back to the direct flash technology.

So then, Wamsi, I do think, just to answer your question specifically around our product gross margins. Look, I think the pricing environment, clearly, we're seeing heightened competitiveness around the pricing environment probably not a lot different than what we've seen historically. That's always been an area where our competitors compete with us. But even despite that, you saw the favorability in product gross margins, which again, is the testament to what Rob and Charlie were talking about specific to our direct to flash management advantages, which, again, I think are sustaining despite what the competition will do from a pricing competitiveness perspective.

Our next question comes from Krish Sankar of Cowen.

This is Eddy for Krish on Cowen. And congrats on strong results. Going back to the AI question, of course. Just at a high level, what is the predominant storage solution for AI today? Is it hybrid storage or flash storage, right, when you go to our customer already working on AI applications, what kind of systems do you usually replace a competing old flash solution or a hybrid story system?

Yes, I'll take that. Well, AI systems are typically they're greenfield. So we're not generally replacing. What we are competing with are solely all-flash systems. Hard systems just can't provide the kind of performance necessary for a sophisticated AI environment. . Of course, you still have hard disk systems in there for some analytics environments, where the performance is not generally as required. But for anything that's machine learning or real-time AI-oriented, it's only all-flash systems, and we compete on the basis largely of our FlashBlade product, which has been in place for 5 years now and now augmented by the latest generation FlashBlade.

Yes, Krish, this is Rob, just to add on to that. As Charlie said, AI earlier in cycle, generally in the training environments are net new and all-flash. I think the broader brownfield opportunity we see is, hey, so what are the large corpuses of data that enterprises have been collecting for sometimes decades. They've been throwing in the corner on hard disk-based systems that have generally been very, very cold and haven't had a need to access that data. Well, now with AI technology, there's now a demand to apply AI or AI applications to those large data sets. Well, now all of a sudden, there's large pools of data need to be accessible. They need to be to a degree of performance. And I think that's where we see a tremendous opportunity for us with especially in our FlashArray//C line.

Our next question comes from Shannon Cross of Credit Suisse.

I wanted to ask about operating income and margin. Given the outperformance this quarter, I know you didn't really change your guidance for the full year. But I'm wondering, assuming there may be some upside, are there areas that you would look to invest further? Or is this something where if revenue upside grew, we should expect perhaps greater than 15% operating margins through the year given the opportunity for leverage?

Yes, Shannon, I'll take this and let Charlie comment as well. But yes, pleased with, obviously, our Q1 results, including better than expectations, both on top line as well as operating profits. We are continuing to be disciplined in terms of spending, really focusing on key areas and expanding sales capacity and that focus remains. It's not changed from how we're thinking about Q1. So when we look at Q2, we're pleased with our guide. You see the expansion sequentially from Q1 to Q2 in terms of our operating profit and again, reiterating the 15% operating margin for the year, which we feel comfortable with. Charlie, any other commentary you'd have?

Yes, it's a bit early in the year to be speculating, I think, on this topic. What I would say is we think 15% really represents the best compromise, if you will, between continued growth and profitability. And we continue to invest in growth overall as a company. If we -- I would say that that's where our mindset is at the moment, but it's another long 3 quarters ahead of us. So we may -- we'll look downstream before we update you on that.

Our next question comes from Sidney Ho of Deutsche Bank.

Great. I have a question on subscription revenue. So your subscription ARR and revenue has grown pretty consistently 30% a year. Are there any risk that growth will start to slow down over the next few quarters when product sales are actually going through a correction in the last quarter and maybe the next couple of quarters?

Yes, it's a great question. And look, we don't specifically guide to subscription ARR, I really do view this metric as important in measuring the overall health of our subscription businesses. And look, we've stated back in fiscal 2022 that our 3-year CAGR expectations for subscription ARR would be around 30%, and we're tracking nicely to that expectation. . Like we've noted in Q1, we saw just outstanding strength of our Evergreen//One offering. And obviously, that's really offset any reductions you might have on other Evergreen offerings that might be attached to CapEx sales. So look, I think we're continuing to see strength in terms of our subscription ARR growth. The value of Evergreen is really resonating with our customers due to flexibility that these offerings provide. And of course, there's incredible value for customers in being able to use critical data storage infrastructure that frankly stays modernized and you don't have to pull out and refresh. So therefore, yes, I think we're continuing to see strong subscription growth and ARR growth.

Yes. I tend to think of this number as a stabilizing element in our overall performance as a company and that when the economy slows down, customers are much more oriented towards these as-a-service offerings to be able to save on cash outlays and CapEx. And when the economy is strong and they move to CapEx, we get it with the Evergreen attached subscription. So I think there's some balance there and vary a little bit, but I think it's going to be -- it's a fairly stable number.

Our next question comes from Jason Ader from William Blair.

My question is on Portworx. I haven't talked about that 1 in a bit. So I'd love to hear thoughts on, I guess, 2 years in, something like that, how that product is doing? And just what are some of the dynamics out there, some of the puts and takes relative to the point in time when you acquired the asset?

Yes. Portworx had a good quarter, so we're very pleased with the progress overall of Portworx. I would say that the the enterprise market for cloud native applications for stateful cloud-native applications has probably progressed a bit slower in the last year than we had expected early on. . But our expectation is that 5 to 10 years from now, all applications will be designed in a cloud native with containers and Kubernetes. So we're very confident about the future. We remain best-in-class product in that area, according to numerous analyst reports as well as we track sales of competitive products. We're #1 in that space, and we expect that to continue. So overall fees, maybe the market is a little bit slower this past year than we might have expected, but overall, very bullish on the segment.

Yes. Jason, this is Rob. Let me just add a few thoughts to that. As Charlie mentioned, we saw a strong quarter from Portworx and I would call it, in particular, strength in seeing customers expand with us. And I think that to a degree, this is natural as we see this idea of platform engineering, the evolution of DevOps really starting to take hold. We believe this is driving more customers to look for -- look to Portworx for really the complete enterprise and scale-ready solutions for their cloud native applications. I'll also point out that I think we're starting to hear stories back from customers that, hey, Portworx is saving them a lot of money. So whether that's optimizing -- helping them optimize their virtualization strategy, optimize their cloud storage and compute cost and spend or just speeding up their overall time to market. And so as we've discussed with other elements of the portfolio, I think there is definitely a focus across the board on value that customers are essentially solutions -- our solutions that are able to save customers money in this environment.

Our next question comes from Nehal Chokshi of Northland Capital Markets.

Yes. Congrats on a strong quarter. I wanted to ask about the media and win that you guys cited last week in a press release and specifically you guys the voice recognition modeling cycle in 6 months to 2 weeks, which is effectively been an order of improvement. So a few questions on this presser. First, is this app degenerative AI study and then is the voice recognition modeling another way of saying basically the training period? And then finally, is the typical level of benefits to customers are seeing, i.e., in order of magnitude performance improvement and it sounds like the typical structure that's being completed with is indeed all-flash arrays with the AI if you can sell to.

Multipart question, Nehal. We're going to try to consider it a 1 question and condense the answer. Please go ahead, Charlie.

Yes. As Nehal mentioned, we actually had 2 press releases last quarter associated with -- or recently, I should say, recently with first quarter wins in the AI space, Median and Greater labs. These were releases that these organizations themselves put out. So we're very pleased to see it. In each case, dramatic improvements in the overall speed of training of their various environments. I'm going to have Rob speak in more detail on Media end, which was the folks of your question.

Nihal, I think what medias and saw is not atypical from customers that are scaling their AI training environments, which is as they started down the AI path, they're doing a lot of training on smaller data sets. These data sets might be sitting directly on the GPU servers. And so at small scale, that works really well. The issue, of course, is, as you know, to produce very good results with AI, you need to apply it to very, very large sets of data and immediately, what customers run into is this challenge of, hey, how do I get a subset of my large, large pool of data, which might be sitting on cold systems over to my GPU servers so I can go and crunch on them and feed them GPUs. And so I think what they saw was again, not atypical, they were spending a ton of time waiting for data to move back and forth between these disparate systems. What Pure FlashBlade was able to do for them is essentially collapse those systems, right, and allow them to train directly off of other shared storage, thus removing a not just long manual steps, but just reducing the overall time to training. And to your first part -- or I guess, maybe the third part of your multipart question. These types of results are, I would say, are not atypical as AI projects start to scale beyond what they're able to achieve with a small-scale infrastructure.

Our next question comes from Tom Blakey of KeyBanc Capital Markets.

I think I'm going to go back to Portworx as well actually, just the DB announcement that you had in the press release. Just wondering what the driving force there just in terms of market demand was there, where does incrementally bring to Pure and the Portworx the data services platform. any kind of update on details you've relayed into the partnership is Mongo serving in the channel? Is this consumption based, et cetera? And if I could squeeze one more in, I don't know if I heard the answer to question about metal in the fiscal '24 guide from even that would be helpful?

I think we're going to need to take the Meta question. We're running out of time, and we have quite a few people left in the queue. So I'm sorry everybody, but we do need to stick to our policy.

Yes. And then, Tom, on Meta, our annual guide continues to exclude new Meta orders. And in particular, future phases, i.e., Phases three and four of the RC environment. So no changes from our annual guide, which again, we have reiterated this quarter.

Our next question comes from Simon Leopold of Raymond James.

I was interested in Charlie's comment about the demise of heart several years out. And just wanted to see what your thinking is or your take on the hard disk technology known as heat-assisted magnetic recording or HAMR, whether that's a competitive threat or how you think about that in the land?

Yes. Absolutely, Simon. Well, look, the density of hard drives will continue to increase. That's a logarithmic curve that hasn't failed for over 40 years. Unfortunately, what's not going to increase is the I/O speed on and off of these disks, which is now becoming more important. What's also not going to change is the overall weight and failure ratios of these devices. And as systems become larger and more dense flash is just accelerating its performance curve beyond hard disk at an amazing rate. So when we say that we're able to replace hard systems that exist today at 1/10 the space power and cooling, even with these new hard disk technology, we're still 5x better, and we'll accelerate beyond that. So I think it's going to be very hard for the hard just to keep up. One last thing to remember is the last refuge for hard disks now is in the secondary and tertiary tier. And now we're able to reach price parity with them at a procurement cost and yet have much lower total cost of ownership and be smaller and be more reliable. So that's the -- there's no other markets that are going to hold revenue for hard disks that flash won't penetrate. And what that means is just lesser revenue and therefore, lesser investment in ongoing development of hard disks. That's also going to be a problem for the vendors. So it's unfortunate I don't hold any malice. But similar to markets in the past, you're just -- when these transitions take place, CDs over vinyl or DVDs over VHS, there's just no stopping progress.

We're going to actually run over by a couple of minutes. We're going to try to get in at these three more questions if we could.

Our next question comes from Aaron Rakers of Wells Fargo.

This is Jake on for Aaron. I was just hoping you could talk a little bit about your views on component pricing for the rest of the year and maybe its effect on flash bull IDs ramp?

Yes. I would say that we're seeing -- obviously, we've seen a significant drop in flash pricing over the last several quarters. We're expecting the same that you're reading in analyst reports, we're expecting that to stabilize through the remainder of the year. But our economics on FlashBlade//E are really strong and compelling and -- we believe that's -- as we said, we're offering it now at the same price as hard disk systems for nearline, and we expect our improvements in density regardless of flash pricing to allow us to accelerate as we go into next year, meaning that we can penetrate ever deeper into lower and lower cost tiers of disk while maintaining the kind of margins that we expect company. So I hope that answers your question.

Our next question comes from David Vogt of UBS.

Great. Charlie, I just wanted to go back to your product road map and how you're seeing to develop going forward. I know you're shipping solutions up to 40 terabytes today. But given the exponential -- the likely exponential growth in data going forward, can you kind of talk through how you're thinking about scaling your business going forward as some of the HDD guys are talking about 50 terabytes and up to 100 terabyte drives getting deployed over the next couple of years. Just would love to get your thoughts on how you're thinking about your product road map?

You bet. You bet. And thank you for the question. Well, we're expecting to deliver a 75 terabyte SSD DFM, right? We provide direct flash modules. We'll provide 75 this year. We expect that to double next year and to double again by the year after that. So it's an incredibly it's not -- we believe that, that's not an ambitious road map. We believe that's eminently doable. And there's -- if you look at the road maps for hard disk vendors, they can't get close to that and certainly not at the same power weight power size space cooling envelope that we'll be fitting in

Yes. And this is Rob, just to add on to that. To Charlie's point, that road map is -- we have very high confidence in it. It's based on effectively existing technology. We don't -- no new physics needs to be invented to make that happen. And I think that's a road map that, again, it's my earlier discussion, I think lastly distances us from the hard disk road maps, but as well the SSD manufacturers, right? We just do not believe that the SSDs are going to be able to keep pace with where we're headed. And that's one of the reasons why we're so bullish about our advantage here. .

Let's take one more question, please. So this will be the last question.

Our final question for today comes from Eric Martinuzzi of Lake Street.

Yes. Curious on the CapEx. I'm looking at it year-earlier there for Q1 versus Q1 a year ago. looks like we're up about $18 million or so. I understand you called out the capitalized software investments up a couple of -- what else is driving that increased CapEx spend here in Q1 versus a year ago? And then what are you expecting for Q2?

Great question. And really, there's 2 drivers for that. It's really around our test equipment for new product releases. And obviously, we've come out with FlashBlade//E and we've got a lot of stuff in the work. So we've got some test equipment investments associated with that and then we're moving into our new headquarters. And so we've got some additional CapEx associated with that as well, but again, when we look at it in terms of our CapEx rate, we'll see a little bump this year against as a percentage of revenue, but around 6% to 7% is what we're thinking.

Thank you, Eric. Before we conclude, Charlie has a few comments to make. .