All right, looks like folks are still trickling in, but maybe in interest of time, let's get started. Good afternoon, good evening, everyone, depending on where you are. Really delighted to host this webinar with the team at NetApp. And really, the focus of the webinar is, you know, how is NetApp poised to benefit from the broader enterprise AI deployment and how they are differentiated versus what seems to be an ever-evolving competitive landscape. A couple of housekeeping items before I introduce the speakers. We'll aim to keep this webinar to around 50 minutes. We'll have the executives from NetApp kick this off. And, you know, actually, we'll start this with Kris doing some safe harbor disclosures. We'll get into some Q&A from there. But I would encourage the intent of this is to be interactive.
So if folks in the group, and it's quite a large group here, have any questions, feel free to either email them to me, or better yet, you can put them into the Q&A box at the bottom of your screen, and we'll weave that into our discussion over the next 50 minutes, with the team at NetApp. So with that, really delighted to have with us Kris Newton, who's the VP of Investor Relations at NetApp. In addition, we have Russell Fishman, who's the Senior Director of AI Solutions, at the company. Russell is responsible for solutions, product management, including NetApp's AI solutions portfolio, and he's been at this function for the last 4.5 years, but he's really been at NetApp for the last decade, plus.
We also have Hoseb with us, who goes by, I guess, only his first name, who is the Senior Director of Global AI, Data Analytics, Sales, and Go-to-Market with responsibilities and oversight of NetApp's AI sales and go-to-market strategy. You know, Hoseb has been in this function for the last five years across multiple geographies, starting off in Dubai and in the U.S. currently. And so leave it with those introductions. I'll pause. Kris, I'll let you go through your safe harbor disclosures before we get into the questions.
All right. Thanks, Amit, and welcome everyone. Today's discussion may include forward-looking statements regarding NetApp's future performance, which are subject to risk and uncertainty. Actual results may differ materially from statements made today for a variety of reasons, as described in our most recent 10-K and 10-Q filings with the SEC and available on our website at netapp.com. We disclaim any obligation to update information in any forward-looking statement for any reason. That said, I will remind everyone that this is much more focused on technology and the use cases, and the market evolution as we see it for AI, rather than a financial update. Back to you, Amit.
Perfect. Thank you, Kris. You know, I guess, you know, Russell said, maybe just to start off with, right, spend a few minutes just talking about the history of where, you know, NetApp has played when it comes to AI, and what has the team been working on over the past, call it, 5, 6 years around this, you know, AI development. And really, the intent of this is just to understand kind of the history of the company, where on the AI side, and then what do you see as bringing you forward.
Sure, Amit, let me take a step on that. So you know, we all know that this whole AI wave has been around for multiple years, right? So in 2016, NVIDIA obviously announced the DGX-1. Before that, there were other people doing a ton of different analytics type of workloads. But really, the AI wave started with NVIDIA announcing the DGX-1. And obviously, after that, NetApp saw an opportunity here, especially in those areas where we call the deep learning at that time and predictive AI. We came up with the reference architecture with our ONTAP storage with DGX-1. We called it ONTAP AI in 2018, and since then, we started adding a lot of customers in different verticals, you know, the strongest being the healthcare, public sector, et cetera.
And that's where our mission and journey with AI started. That's where I started, taking over the mission, moved to U.S. to build the team, and also our product and engineering teams built that references with NVIDIA and kept going on, right? Till we saw the wave of obviously GenAI with with ChatGPT and OpenAI. And and we continued the innovation. We our reference architectures kept growing with NVIDIA. We added the SuperPODs three years ago, then obviously the A100s, the H100s, et cetera, with the NeMo Retriever right now. So we kept innovating in the past six years, adding customers, as I said, in different sectors. Back then, when we were doing deep learning, customers were leveraging our technology to build the models, even sometimes we used to call machine learning.
So, you know, one of our first customers was a large healthcare provider in United Kingdom, who basically deployed ONTAP with DGX-2 from NVIDIA at that time to build models to feed these CT scan images into those models, and, you know, predict whether a tumor is benign or malicious, et cetera, et cetera, right? So, till today, where we have, for example, recently closed a large SuperPOD opportunity in one of the leading genomics and healthcare companies, where they will be building models that will be trained on the world's largest human patient data, right? So, this has been the progress that we had in the past six years, whether on the sales side, and the product side, and the engineering side.
I'll just add to that. I think one of the big changes that we've seen is this inflection point in the last 12-18 months, where AI has moved from being solely the domain of companies that are looking to innovate and get ahead, to something that is seen as a necessity to stay in touch with their competition and their industries. And to that end, you know, one big change that NetApp has made in our fiscal year, this current fiscal year, is that we've moved from purely a specialist sales motion to one that is far broader, that touches every single one of our salesforce. And I think that's just a reflection of the interest level amongst our customer base, amongst all customers, to want to engage with NetApp around AI. Perfect.
You know, I think it's a great point, which is this is something starting to get more wide, you know, wider in terms of deployment beyond just the four, five, six hyperscalers. You know, I think one of the questions I get asked, I'm sure a lot of folks, you folks get asked a fair bit, which is, you know, what is the, quote-unquote, "AI opportunity" for NetApp going forward, right? How do you kind of put dimensions around that would be really helpful. You know, maybe as you answer that, I would love to understand that, in three buckets, if you may, which is, you know, one, kind of the incumbent install base, how does that kind of stack up?
Secondly, the next new infrastructure builds, and then third, you know, there's a vast amount of data that actually needs to be prepped, for enterprises for the AI use cases, right? So maybe just talk about, you know, what is the AI opportunity? How do you size that up to NetApp as you go forward?
Yeah, I would love to first just start by characterizing the workload a little bit on the AI side. So you've got definitely the training workload, which is basically building those models where data gets fed into the GPUs, and then it trains the model. And obviously, you've got the inferencing piece, which basically putting those models into production, and then now you're inferring on those models. And then obviously, on top of that, you've got techniques like fine-tuning and RAG. So once we, you know, segment those, we will better understand the opportunity for us. When it comes to model training, let's start with that. Model training is actually something that we have been engaged, as I mentioned, since the early days of the DGX inception. Also, in addition to that, a GPU-based, other OEM-based servers, right?
We have customers today who have deployed servers from different vendors with GPUs in them, but they relied on NetApp technology to feed those GPUs with their data and train those models on. So I can tell you that NetApp has been very instrumental in that space. The opportunity for us over there, whether a customer is building a separate cluster to train those models or they are leveraging their currently existing data that sits on ONTAP technologies or, or NetApp technologies, whether it's in the cloud or on premises, we have built all those integration points to be able for those customers to feed that data into the GPU clusters. So, I wanna repeat that in the training, we've been really, really helping a lot of enterprise customers.
We have customers all the way from 2 DGXs, for example, into more than 80 DGXs in one of the largest model training clusters that we have today, in one of the financial institutions. So in model training, we see a good opportunity for us. We see that opportunity growing, but more than that, if you come to the RAG and the fine-tuning, that is where we also see a tremendous, tremendous opportunity for NetApp, which actually sparked a couple of months ago, as Russell mentioned. You know, for those folks on the call, RAG is basically a technology which is called Retrieval Augmented Generation, is where you don't need to retrain those models using huge clusters, but you can deploy those pre-trained models but make it relevant to your businesses, make it updated.
Because traditionally, those models would be trained on a year-old data, right? So that model gets into production, and then if you use technologies like, like RAG, and that becomes very important. Now, you would be asking, where is the storage in this RAG technology? The storage comes actually. The RAG technology requires a pre-existing data to be converted into numbers, into numerics, that the models would understand better. You know, the industry, analysts, a couple of them already out there saying that that process will actually generate almost from 1x or 100% more data to almost 10x of the data, because those embeddings would generate, would need more, more, more numerics to be stored on. So that is the opportunity.
Now, that is an opportunity a lot for our install base, and this is where NetApp is really well-positioned in the market. Being in the market for more than 20 years, storing that files for our customers, enterprise customers, for almost 20 years, gives us the ability now to transform those data that is sitting on NetApp install base into this embeddings. So that is a huge opportunity for us that we anticipate in the next couple of years, especially in the RAG technology. And obviously, the last piece is the inferencing, right? The inferencing is something that we anticipate, for example, these RAG models, when they start generating new types of documents, PDFs, videos, audios, et cetera. One might be saying, but those would have been generated anyways without the RAG models or without the GenAI.
But I would say the pace of those data being generated is gonna get accelerated. Because if you were to be able to produce a movie a year, now you would be probably able to do two movies a year, three movies a year. Same with documents, right? So those type of new documents that would come out of the inferencing we anticipate that to be also a good business opportunity for us. Now, put a wrap around this is the cloud and the hybrid cloud story we have, because we believe that AI is a true hybrid workload. So that means it is an opportunity for us both on premises as well as in the cloud, as well as in a hybrid model as well.
Yeah, I'll just add to that. So a couple of things. So if you look at how industry analysts look at the amount of unstructured file data that sits on NetApp, it's north of 30%. So we have an outsized share of that market, and that is the field that's driving the current GenAI wave, particularly. So we're extremely well-positioned when it comes to GenAI. But more broadly, you know, my suggestion to the folks on this call is to follow the data. I think that what you're seeing is new entrants coming into this space who don't have any of the data, who inherently have to start by taking data from somewhere else, creating a new copy of that data, managing it.
That's, that's not just expensive, it's complicated to manage. It even causes problems with things like the AI Act recently launched and put in force by the European Union, where the ability to know where your data is and how it's being managed becomes not just a case of best practice, but one that is subject to regulatory and compliance concerns as well. So I think that, you know, NetApp is extraordinarily well-placed from that perspective. So incumbent install base, for sure. What we see is certainly in training environments, these are all net new infrastructure builds. So, so, so all of that business is being in that space. Obviously, when it comes to generative AI, we're talking about taking or augmenting existing environments, so we see that as well.
In terms of just, you know, the amount of data needs to be prepped, you know, we'll talk about that, I think, as we go through this call, and we'll get into a little bit more detail, but it's significant. As Hossab said, it's also growing, not just in terms of the amount of data that's growing, but the way that data needs to be prepped and worked through, actually generates more data in and of itself to run through techniques like RAG.
That's great. And, yeah, I do want to spend a little bit of time just on RAG and inferencing, what that means for you folks as it goes forward. But, you know, when you think about some of these kind of capacity requirements, if you may, you know, mostly to understand, you know, when do you expect to start seeing benefits from the growing data capacity requirements that some of these AI models have?
Well, it's happening now is the answer to the question. I mean, we're firmly in this gen AI wave, but I'll take a step back and just give you some sort of industry numbers, if you will, on sort of what this data growth looks like. So, Hoseb mentioned this technique called RAG, or Retrieval Augmented Generation, which is really the leading technique used to augment foundational models with context for a particular company's data environment. And, as part of that process, that data goes through a process that generates these things called vector embeddings, and these embeddings get stored in a vector database. That vector database then gets connected to the foundational model at the point of prompting.
So when a user asks a question, it is able to use the vector database to connect to underlying data sources. I kind of think about it like a librarian with a bunch of books behind the librarian. The librarian themselves doesn't know the content of all the books, but knows how to go find that information and how to read those books and get the right information to answer the question. In terms of how much data is generated, the answer is, I think the industry as a whole is continuing to refine its answer in this space. But what we have seen as an industry is somewhere between 4-10x of typical office file environments.
So that means if you have, for example, 1 TB of office file data, that would be, you know, Word, Excel, PowerPoint, PDFs, et cetera, et cetera. The sort of content that typically gets used in, for example, an enterprise knowledge management use case, we would be seeing somewhere between 4-10 TB of additional data generated, as part of that RAG process, and that, that data doesn't disappear. It stays there. It actually continues to get updated as the source data changes or expands. Interestingly, this data also, with current techniques, doesn't seem to be compressible in any way. So that actually translates pretty clearly into an increase in raw capacity requirements. But I've given you, you know, one example. There are other examples.
If we look at another common example, which is using code bases for retrieval augmented generation, we're seeing 200x data increases. So it really depends on the type of data, and I think as the RAG market gets more mature, I think the industry will continue to evolve standard metrics around this. But clearly there's a significant opportunity there. In terms of when, you know, we're seeing a huge amount of growth and interest in gen AI. One of the reasons why I mentioned before that we've really engaged all of our NetApp's general sales force, for example, in engaging with customers around this, is because the appetite is there. Gen AI is probably, I would argue, one of the most accessible forms of AI.
It, in many cases, moves directly to a value phase from a customer ROI perspective, and so that increases the level of interest from customers who may not have had the sophistication to go build models from scratch. So we really think that wave is happening now, and, you know, it's something that we expect NetApp and other, you know, folks in the industry to gain benefit from in the next 12-18 months.
Got it. You know, as you start to think about all this, right, how should we think about the timeline of how data is stored through the AI life cycle in its entirety, right? Like, what happens to the new data that's created? You touched a little bit on, hey, you can't compress this and everything else, but, you know, Will there be different performance requirements depending on where you are in the life cycle, you know, be that training, tuning, or RAG or inferencing, for example?
Yes, there's a lot to unpack there, Amit. I would say that one thing that we have seen as an industry is that because data through a typical life cycle for data and AI has been highly siloed, if folks don't use solutions like the ones that NetApp provides, but if you allow each of the different personas that typically are involved in making AI real for a company to do their own thing, you could end up with as many as six redundant copies of the same piece of data through that life cycle. So, there is. It's really interesting in terms of an optimization opportunity that obviously NetApp is very well placed to take advantage of.
I think what's kind of embedded in what you just asked, which is around the performance requirements, the reality is, there is a significant variation in the performance requirements. In fact, the workloads we kind of talk about AI as a singular workload. In reality, it's lots of different workloads with lots of different use cases, each one of those use cases with their own performance characteristics. So what we're seeing is customers are gonna be best served by a truly flexible solution that can deal with data at all stages of that pipeline.
So whether it be at the data unification phase or the prep phase, or the training phase, or the model management phase, or the inferencing phase, or the feedback loop phase, each one of those is really a different set of performance requirements, is a different type of characteristics. And one of the things that NetApp's really focused on and, you know, is taking the amazing ONTAP leading storage operating system and applying it in lots of different ways through that process.
And I'll just say one other thing that's kind of interesting here, and I think we, I think, well, again, this is one of the things we'll probably get into a little bit more as we go through this, Amit, is that, you know, if we think about all the workloads NetApp has seen over the last 20 or 30 years, I would say that AI is probably the most hybrid workload we've ever seen. And, actually, if you hear other folks in the industry talk about the AI workload solely through the lens of cloud or on-premises, they're doing that because of their portfolio.
But if you actually go and talk to customers, and I—when I say customers, I don't just mean lines of business, but data scientists and data engineers, what you'll hear is that that flexibility and the fluidity that they need, to move and manage data, both on-prem cloud, through service providers and other places, bringing data to GPUs and GPUs and AI to the data, you know, they are, are gonna be very well served by the strategy that, that NetApp is pursuing in this space.
Got it. You know, I guess this would be a good time for you folks to talk a little bit about, you know, what do you see the competitive landscape as right now? And, you know, maybe just talk about, you know, there are a lot of companies that are, I guess, you know, lack of a better word, AI-washing their narrative. In fact, who do you see really in these customer deals that you engage with your customers, you know, who else are they looking at, looking for RFPs, for example? And maybe also contrast, how has this landscape changed from a historical perspective for you folks?
Listen, I'll, I'll give you some facts, and I'll let the audience decide who is AI-washing or not. But I think it is very, it is very obvious that we need to, we need to clarify what is being identified as an AI customer today, right? Especially in the storage world. If we want to easily call someone just using data to do some dashboards and say they're doing AI, we can call that out day and night.
But I want to really specify when we say our work in the past six years, and I you heard me say a lot NVIDIA and DGX pods and GPUs, because we truly believe that those are the really use cases, that there will be no argument around that, whether the customers is doing an AI or not, because, you know, we all know GPUs and the purpose that customers are utilizing them. So, the competitive landscape, I mean, I would say, has changed over the past six years. And it changed dramatically. In the past, in 2018, I mentioned there were probably two publicly traded storage vendors. They were certified with NVIDIA. So you could see us compete with another vendor in that specific deals, and it was the market wasn't hot as it is today.
So you could basically identify those customers as a handful ones. As we grew in the market, obviously, COVID hit, right? And priorities have shifted. If you look at NetApp's history, we were probably the only ones from a storage standpoint who kept believing that this market is gonna grow, and we kept investing in this space. We kept our teams together, including myself and our specialist team, because we knew this market is gonna grow. And in the time of 2019 till 2021, 2022, we were probably taking customers left and right. We didn't see much competition coming in. Priorities have shifted in other vendors.
But after 2022, when SuperPOD started becoming more and more important in this space, obviously, a push from NVIDIA came on that front as well. We started seeing HPC type of storage vendors come into the play in our deals, right? So we started seeing, I would say, a healthy competition on that front from vendors, also not public, also startups get into the play. But from an enterprise-level storage player, we don't see that much of competition in our deals. Now, you have to be very careful here. Again, very specifically talking about DGX opportunities, GPU opportunities. We also compete with other server vendors who also tend to provide storage, right? So that's a natural competition over there, too, where a customer might buy a GPU from a server vendor and a storage.
But I as I mentioned in the past, we've also were able to attach our storage to customers who bought GPUs from servers, from different types of vendors who also provide a storage. So we've also seen some competition over there. But if you look at NetApp, I think we are the ones that provide that end-to-end, as Russell mentioned, portfolio, that could cater customers' requirements, whether they are building a two DGX cluster for training, whether they are using cloud, like, you know, technologies like SageMaker and others, and their data is in the cloud, that they would love to feed it to those GPUs for training or other techniques like RAG. Or if they are building clusters like Superpods, then NetApp has a solution for that.
I'll just add the one thing that separates us out, I think from our competition in a competitive landscape, is that we aren't competing with the hyperscalers. I think if you go and look at the other folks in this space, especially the ones that like to talk about themselves as AI specialists, by the way, if you ever hear that, NetApp is a specialist in AI. It's just that we don't only specialize in AI. The reality is that we don't compete with hyperscalers, we partner with them, and that is based on a legacy of building out these 1P first-party services, cloud storage services at each of the three major hyperscalers.
And seeing, you know, integration between those services and the native hyperscaler solutions, some of which actually just happened to be announced today. So, you know, that puts us in a very different position. And it also means that, you know, again, back to that idea of fluidity, the flexibility that we offer the actual AI practitioners in managing AI workflows in a hybrid manner, you know, we're not trying to force customers into one modality or another. We can support whatever they want, I think. So as well as supporting best performance requirements, we're also able to support however a customer wants to construct their AI data pipeline.
Got it. I have a question, a couple of questions actually from folks in the call here. You know, one of them is just, it's on the competitive landscape. Is there a different set of customers or a different set of competition you see in the cloud side versus hybrid? And, really, the question, I think, is intended around VAST Data systems and where do you see them play versus not. Maybe we'll start with that, and there's one more after that.
Yeah, you know, I think where we see our value is in the enterprise data level management capabilities. We see a lot of customers, especially on the enterprise side, the large pharmaceuticals, the large, financial institutions, where they really care about their data being secured. They really care about their data being managed efficiently, where they have been relying on NetApp technology for many, many years, and not only price performance density, right? That is also another area that we really need to be very, very, careful about is, hey, there will be customers who are really looking at price density performance, and then obviously us all being certified with NVIDIA, which means we have met the requirements to, work- to serve those workloads. But where our value comes in, as Russell mentioned, is this workload being hybrid.
I was talking to one of the financial institutions last week who happens to have their data sitting in the three hyperscalers as well as on premises. When they wanted to create this foundation for their data to be served to these GPUs wherever they are, they really like the strategy that we presented in terms of creating that data fabric. Our being natively in the three hyperscalers literally differentiates us from any other storage vendor out there in the market. If you wanna be only on price performance density, you can just put 10 competitors next to each other, and they all can compete on the price and the performance.
But our value, again, I repeat, comes from enterprise-level support, enterprise-level security, enterprise-level data management, our history in the market for almost 25+ years, and also our native integration with the hyperscaler consoles in the major three hyperscalers.
I'll just, I'll just add to that. You know, performance is a requirement, but it's table stakes. If you look at the way that NVIDIA has gone about talking about performance when it comes to their relationship with storage vendors, they have spent a lot of time and effort creating validation and certification environments that essentially level the playing field from a performance perspective. You either make, you either make meet the requirements or you don't. If you meet the requirements, then let's move on, and certainly NetApp meets the requirements across all of these solutions, like SuperPOD and BasePOD, et cetera, et cetera. I think, you know, the other thing I would add is, you know, you don't get fooled between the difference between having a cloud offer and having a native 1P cloud offer.
The way that we get treated as an OEM, essentially, to the hyperscalers in the delivery of our first-party ONTAP-powered cloud offers is completely different to the way that they treat the marketplace offers that some of our competition rely on. Certainly, when it comes to integration with their AI services, the hyperscalers are incented to do that work between their services, which, remember, these NetApp-powered services are actually native services in each of the hyperscalers. Again, you know, we just made some announcements specifically today around that with AWS at the AWS summit in New York City demonstrating our commitments jointly with the hyperscalers to integrate our technologies directly into their first-party AI services.
Yeah, and Amit, one more thing. One of our messaging to our install base specifically, and also our customers, is, like, you really don't need a siloed architecture to run this. And if you look at NetApp, right, in the past years, every different technology that came, we try to integrate the solution with our existing tools, with our existing ONTAP, with our existing operating system, which means customers do really not need to buy some new technologies in their data centers that will require them to hire and train new individuals to on... Especially for our install base, which is really, really good compared to others, right? So this is one of our differentiators, is not- we're not trying to create a separate architecture in a silo to run this workloads.
As I said, if we really meet the certifications, then we met the performance requirements already set by NVIDIA, which are very high standards, right? So, so there's no reason why would someone just deploy a different architecture to run the workload that could run on the architecture that they've been using for many years.
You know, I was just talking with NVIDIA. I was hoping you could just expand on your relationship with NVIDIA, and maybe you can frame it around, you know, what enabled NetApp to get SuperPOD certified earlier than, you know, some of your public peers, at least their comparable products? And maybe related to that, you can just talk about, you know, you know, what do you think differentiates NetApp SuperPOD versus, you know, offerings from, you know, your competition right now?
Yeah, I'll take that. We have a very strong relationship with NVIDIA. We're not claiming to be the only one, but we have a very strong relationship. It started back in, as I said, in 2017, 2018, and we continued the building up on that relationship. As a matter of fact, we were very proud to be the only storage vendor being called out on Jensen's keynote at the GTC, the recent event from NVIDIA.
Your question about SuperPOD, it's very important, and thank you for asking that, because that SuperPOD certification, us having the SuperPOD certification and the BasePOD certification, and for the audience on the call, BasePOD certification today, you know, in the past as well, it relied on Ethernet as well as an NFS file system, whereas the SuperPOD certification in the past, again, relied on the parallel file system and an InfiniBand requirement. So they were very distinct, two different architectures. And then in 2020 and onwards, we went through almost 6-8 months of testing with NVIDIA, because of all the requirements they had to be certified. And we were one of the three vendors actually, to be initially certified for SuperPOD.
The reason we had that is because our portfolio was able to do that. Our portfolio was rich enough and had partnerships well enough to be able to pass NVIDIA's SuperPOD certifications early on, so that we can provide our customers the ability for them to choose the architecture and the solution that fits their requirements. So when I walk into a customer, I'll be asking them: "What is your workload? What are you trying to do?" And if they say, "We want to go SuperPOD, that's the architecture that we were recommended by NVIDIA." "Fine, we've got that for you." If they say, "No, we don't have InfiniBand in our data centers, we're gonna rely on BasePODs with Ethernet and all that," we have that as well with our BasePOD certification.
So, that has been a very strong proposition for us, and I can clearly say that we have customers who are leveraging both technologies. We also have a customer who's leveraging both technologies in one time, meaning they all have a SuperPOD for their large language model, you know, training type of workloads, and we also have them using BasePOD to utilize technologies like containers and RAG and NeMo Retriever and the different things from NVIDIA. So, we are very strongly positioned in that space, and that is all because of our rich portfolio that can manage those different type of requirements from the customers.
Yeah, I'll just add that, you know, this is a situation that might surprise the investors to hear actually makes it much easier for what you might describe as a legacy supplier, someone who's been in the industry for a while. Because the market is very fluid, it's changing rapidly. The technologies that are being leveraged are changing. And, you know, it is relatively easy for us at NetApp, with our very broad portfolio of capabilities and rich technological integrations that we've built over time, to quickly adapt to a changing market. Some of the startups, the privately held companies, are having to build everything from scratch every time anything changes.
They have actually an inherent interest in keeping the status quo, which is, you know, kind of ironic in a space like AI, where, you know, the market is moving so quickly.
You know, and then, well, I guess when it comes to enterprises looking to deploy AI, you know, what environment do you think they'll deploy AI for as they go forward? Is it more on the cloud? Is it on-prem, is it colo? And then, you know, maybe related to that, what environment do you think is NetApp best optimized for?
I'll take that. You know, the way we tend to engage with customers, it often starts with a conversation around data gravity. I mean, the... You know, you have to remember that when it comes to AI, AI is nothing without the data. You can build the biggest engine you want, but if you don't have the fuel to power it, nothing really happens. So it all, you know, everything really starts with a data conversation. In fact, what I'll tell you is that the customers that are most mature, that NetApp works with, and Hoss had mentioned some of those industry verticals earlier, those are the ones that have truly recognized how important data is. In conversations with the folks, the customers that really understand AI, the conversation always starts with data.
The conversation with customers that don't really understand AI, don't almost—sometimes don't even recognize that data is the problem, but they recognize very quickly once they start trying to do it, that it becomes data. So, you know, one of the things that we've recognized is that data gravity, that's not just where data has sat, but also where it is generated, is, is really, really important. And, and, you know, one of the reasons we believe this is an inherently a hybrid workload is because data isn't in a neat pile in a single place. It is everywhere, both inside the enterprise, that could be in different data centers and different storage environments. It could be in the colo facilities, it can be in the cloud or any combination of clouds.
We see just obviously, a proliferation of customers going multi-cloud, and that obviously becomes a really complicated thing to manage. So what we recognize is a need for a flexible, fluid, and adaptable environment, that enables customers to quickly, orient both their use of, and deployment and use of GPUs with their, ability to unify, prep, and move data in a very, very seamless way. Now, So I'm kind of not really answering your question, Amit, because in a way, our view is, is that, it needs to be everywhere. There isn't a single place that we need to be. And again, you know, you go talk to customers that have that, have that knowledge and experience in AI, they will tell you absolutely the same thing.
We're of course, in a pretty unique position in that we can benefit regardless of where the data resides. You know, any one of those locations I mentioned, you know, NetApp has, you know, a singular control plane that spans across all of those areas, and a consistent data plane that simplifies that experience for customers. One of the kind of issues in AI, and we kind of simplify it significantly, and not just with one type of data. I mean, I think what we see in AI is a increase in the use of multimodal environments. That means different types of data, whether that be textual or image, audio, video files, you name it. Different types of both static data and streaming data.
You know, those are all the different types of things that we, you know, we need to be able to manage, and we can, right? Whether we're accessing it through file or object, it doesn't really matter to us.
Maybe I'll ask you something, and the question was really like, you know, why is NetApp able to benefit regardless of where the data resides? Because that's a pretty unique proposition for you folks. So maybe just talk about why is that, that's the case, and, you know, how is this differentiating, resonating with your customers? I think, Lockheed Martin, for example, has built an AI center of excellence using NetApp. So maybe just talk about, you know, why is NetApp able to benefit regardless of where the data is stored, and how has that resonated with customers like Lockheed Martin?
Yeah, I'll take the first part, then I'll let Hossab discuss Lockheed. You know, okay, so I mentioned, of course, our ability to do NetApp everywhere, right? So ONTAP everywhere. So whether that's on-prem, the cloud, I think Hossab also touched on the completeness of our portfolio. So the way that we're able to deliver different types of storage environments, focused or optimized for different storage needs, that's really, really important. And I'll point out that, you know, back to the question you asked earlier, Amit, which is all about, you know, how that data is stored throughout the AI life cycle. You know, each one of those phases has different requirements.
So as well as potentially being in different places, it also has different needs, depending on which part of the data pipeline touches that deployment modality. So again, you know, richness of portfolio is part of it. A singular and consistent data plane and a singular and a unified control plane are the ways that we're able to deliver that, both on-prem, cloud and with key service providers. So that puts us in a, you know, super unique position, compared to really the rest of the market. And on Lockheed, Hossab?
Yeah, you know, to tap on what Russell said, a corporation like Lockheed, having different data scientists, multiple data scientists sitting in different parts of the United States as well as in other countries, will definitely having this ability to have the data closer to where the data scientists wanna leverage their workforce. It's not one use case, right? It's not one batch file running for days. It's literally multiple hundreds of data scientists. They built a center of excellence where all these people can actually leverage the infrastructure they built. Now, if you don't have that agility that Russell was talking about from a data perspective, you'll end up with different copies of the data, you will jeopardize the security of the data, the governance of the data, or you won't do the job that you are intended to do.
So that's why they will be relying on, or they are relying on technology like NetApp, in the cloud and on-premises, to have that agility of the data to move around.
I don't want to rabbit hole on this, but I do think there's a really, you know, critical point I want to highlight from what Hossab said: data governance and security. You know, I cannot stress enough how critical that is for our customers. The types of data that is being used in AI environments is the customer's—often the customer's most critical data. That—and when I talk about data that might be subject to regulatory or compliance concerns, things like personally identifiable information or other things that are subject to regulatory regimes, in various jurisdictions, but also just commercially extremely sensitive information.
Information that, you know, would sort of spark a shareholder revolt if it went into the wrong hands, because it really a lot of the value in the company is based on, you know, customer buying trends, for example. So, you know, being able to have a consistent approach to data governance and security throughout that data life cycle is really, really important to our customers, certainly to CDOs that we talk to. Again, you know, having a single vendor view that can extend beyond, you know, throughout that life cycle through all of those deployment modalities and through all the different data types and different phases of that life cycle, significantly reduces the risk.
Capabilities like NetApp's leading anti-ransomware protection, backed with our anti-Ransomware Recovery Guarantee , demonstrates NetApp's commitment to protecting this incredibly important data through all phases of its life cycle.
Oh, that, that's really helpful. And then, you know, one of the big differentiators I always thought about NetApp has been, if you compare to your peers, it's this first-party integrated position that you have across all three big cloud service providers. You know, how, how is that proving out to be a differentiator as enterprises look to deploy AI? And then we can also just talk about, you know, what work are you folks doing specifically with these hyperscalers to integrate with their own AI offerings that each one of them have right now?
Yeah. So I'll start by saying that, it's important to note that many, many of our customers get started on the cloud, and there's a lot of different reasons for that. Obviously, it's easy to get started on the cloud, so it's relatively straightforward. But also just things like the availability of GPUs, and not being just the availability, but the cost of getting started with your own AI environment is high. And it means that customers are often looking, you know, are we sure that we wanna go do this? Often, that means that they go and start in the cloud, right? So, you know, that's the first thing.
Even customers that have decided in the long run to move some or more of their AI environment on-prem for cost and efficiency reasons continue to leverage the cloud to balance their capacity requirements, to find that optimal combination of highly utilized GPUs on-prem with bursting capability in the cloud. So, you know, the cloud is not going away. The customers start there; they often end up in a hybrid modality. You know, within the cloud, one of the things that the clouds do very well is provide these first-party services.
So that includes things, everything from the development side of AI, like MLOps, which is more on the training side, to the inferencing side, probably most famously through their generative AI services, across each of the three hyperscalers. And so, you know, again, as I mentioned earlier, you know, we are at a super unique position to be integrated into those. Earlier this year, we announced an integration between the Google Cloud NetApp Volumes service, so Google's first-party storage service built on NetApp ONTAP technology, integrated into their Vertex AI platform, which does a number of things, but in this case, was focused around GenAI use cases.
And what we've seen, you know, interesting, you know, significant take-up of that, of that solution that's in public preview, enabling customers to quickly extend their existing data environment to go sit next to the public Vertex AI solution in a way that protects their data privacy and doesn't expose their corporate data into public models. Very interesting. Today, as it happens, we happen to announce the extension of that GenAI Toolkit into Azure with ANF. We also announced with AWS our BlueXP Workload Factory, it directly integrating our solutions into Bedrock, for example. Bedrock being the GenAI development environment, the AWS service that AWS offers, and that's now directly integrated with our FSxN solution.
So that's Amazon's FSx first-party storage service, again, built on NetApp ONTAP technology. So we have a, you know, again, pretty unique position in that regard, and it is a. It's a huge differentiator because NetApp can uniquely capture this data where it starts, rather than just having to capture when a customer decides to extend that environment on premises. And rather than having to say to a customer, "Hey, whatever you're doing in the cloud, you have to do something very different on premises," we can actually make that experience extremely consistent and flexible and fluid.
I would add that, you know, when it comes to, you know, where data scientists and data engineers, the AI practitioners that actually make AI real for customers, actually, where they focus on, the one thing I can tell you is they really don't care about infrastructure, and they really don't care about storage. They do care about data set management, and that's something that we really focus on, which is exposing our storage value in a way that's easily consumable by the AI practitioners in a way that makes their lives easier, faster, and more effective.
Yeah, I mean, I just wanna add one more thing. From a customer standpoint, this basically simplifies their AI journey, right? Because, there's a lot of complexity in the data, and if you tie this to the first questions that you asked about the workloads, RAG or fine-tuning is not necessarily you just spin up a new environment and put a new data in there, right? It's basically relies on the data that you had, and you've been storing it from different customers, from internal customers, et cetera, that you have. So you would need those data that have been stored, to convert them into embeddings. So if it was just a new environment and we're starting fresh, I would tell you this probably wouldn't matter much.
If you're really like the technology in one of the cloud providers and your data sits in another cloud provider, but you need to bring that so that you can make these models ready for your business, then you would love this technology because we will enable that for our customers. We know, we've talked about installed base. Obviously, this is one differentiator for NetApp that helps us also get net new customers, right? Because customers who haven't had any on-premises environments that have been doing AI or any other workload in the cloud, now they wanna move back on-premises because of different reasons. You can take this as a great differentiator for us, and we've been doing that, of getting net new customers from this angle.
Got it. I know we're coming up on our time, but I have a question from one of the folks on the group here, so I'll ask that. You know, can you talk about, you know, what is driving the willingness of enterprises to transform their existing data into vector embedding? And, you know, maybe a little bit of the spirit of this is, why would an enterprise take on the cost to transform existing data into these vector embeddings right now? Maybe just touch on that dynamic a bit.
Yeah. I mean, so, actually, going down that route is a significant savings for companies that are thinking about getting into generative AI, and I'll explain why. The alternative is to go down the route of training your own model. And let's be clear, most customers, actually, I won't say most. The vast, vast, vast majority of enterprises do not have the data or the wherewithal to train their own foundational models. They are essentially stuck with leveraging foundational models that are either publicly available or available on a subscription basis.
If you're, if you're doing that, then, you know, you're saving the cost of not having to build your own foundational model, but you still need to make it contextually relevant to your own data sources. And that's where techniques like RAG and also something called quantization come into play, which enables a customer to take information and foundational model kind of narrow it down to the sort of areas of knowledge that actually is needed for the task it has in hand. For example, if you're building a customer services chatbot, you obviously need that customer services chatbot to know how to deal with logistics and deal with customers. It probably doesn't need to know, you know, results from the, you know, baseball World Series for the last 10 years. And so, you know, quantization removes some of that unnecessary data.
RAG, what RAG is really doing there is, rather than having to retrain a model each time, get the underlying data or policies change, RAG allows you to make those changes without going through the expense of doing the retraining. So actually, you know, RAG, you know, obviously produces data. There's costs associated with that. Even the development of the vector embeddings themselves, you know, can be computationally, you know, expensive, if you will. But compared to actually going through training or retraining, it's actually a relatively minor cost and enables customers to stay very much up to date with the latest information changes in their underlying data sources.
Yeah, Amit, security is another aspect too. So when you look at RAG technology, it basically takes the raw data and converts into to numericals, which basically you're not exposing any of your data. So that's another angle where customers would love to use embeddings and technologies like RAG, so that they are not exposing their IP data. It's another important reason why they would love to, not love, but they would prefer to convert their data into embeddings.
Oh, Amit, you're on mute.
Amit, we lost you.
I mean, it's been four years of doing this. You think this would become a lot smoother right now, which is not. I just say, I know we're coming up on our time, but maybe I'll ask. I'll squeeze one more question in before I turn this back to you folks. Maybe we talk about, you know, how is your go-to-market strategy evolving, you know, to capture the opportunity that comes to enterprises and AI deployment, and, you know, how different is the AI deal, you know, from RFP deployment process versus the traditional deal? We just talk about go-to-market stuff a little bit.
Yeah, I'll take that. Obviously, we have a lot of learning and history in the past 6 years, being in hundreds of opportunities, competing in hundreds of opportunities. The AI deals are not similar to a storage RFP, where a customer wants 100 TB and certain performance, and then you just do 3-4 months of the RFP cycle. Actually, we have noticed that AI opportunity starts from the use case itself, and it goes down into a POC. You have to show the ROI to the customer, so it can take all the way from 9-12 months. Now, that has sped up, obviously in the past year, but it's still. It's a completely. I wouldn't say completely, it is fundamentally different than any other storage.
And as you guys a storage RFP, and as you know, NetApp doesn't only do AI type of workloads. We are also in the SQL databases, Oracle, et cetera, SAP. So, you know, to help our sales force to position ourselves better in these opportunities, we have put together a team that is specialized in this space, that their mission is to go and help our sellers to penetrate into either net new opportunities or net new buyers within an install base. To take that opportunity from its inception all the way to a POC into a production would require a lot of work.
You know, that's where I think we differentiate ourselves also on the market, having this specialized team that are really focused on this type of opportunities, and also having our sellers focus in other areas as well, including our core business, block, file, et cetera, et cetera.
Perfect. I think with that, maybe, you know, I'll stop my questions here. I'll turn the virtual mic back to, you know, Kris, Russell, Hossab, Hoseb to you, see if there are any closing comments, anything we did not touch on that anyone, if you want to flag our way as you think about AI and NetApp's value proposition there.
I think, in closing, we believe we're strongly positioned in this market. We have a unique value proposition that expands from on-premises to the cloud. We have a strong partnership with NVIDIA and other players in the market as well. We have built a foundational customer base in the past six years that we learned a lot. We know this workload. We've built an expertise in-house, and also a dedicated go-to-market as Russell mentioned at the beginning. So, I would say we're very extremely excited about the future, and we're moving forward with it.
I think that's it from us.
Yes.
Thanks, Amit, for hosting us.
Okay.
We really appreciate it.
Thanks a lot. It was a pleasure, and thanks for your time, Hoseb, Russell, and Kris. We'll chat soon. Thank you, everyone.
Thank you.
Thank you.
Thank you. Bye-bye.
Bye.