Good early afternoon, everybody. We have Nutanix today, CEO Rajiv Ramaswami and CFO Rukmini Sivaraman. Welcome back, guys.
Thank you.
Thanks for having us, Matt.
Awesome. All right. Rajiv, let's kick it off with you. A lot has changed around Nutanix over the past year, from the continued disruption of VMware to the progress across external storage, public cloud, AI. When you step back, what feels most different about the opportunity in front of the company today?
Yeah, I think it's important to realize that Nutanix is no longer just a HCI company. That's how we started, that's what got us on the map and brought us to where we were. But today, we are a much broader cloud platform company. What that means is that we can handle a much broader set of customer needs, resulting in a much larger market opportunity for us to go after. Today, the platform itself works on-prem, it works across multiple public clouds. It supports built-in storage with HCI. It supports third-party storage, which makes it much easier for customers to migrate to our platform. We have a full-scale Kubernetes platform. We have a database platform. And increasingly now, we also have an enterprise AI stack that's starting to gain traction with customers as they look at deploying inference. So the platform has become much bigger.
What that means is that if you look at the landscape of what we can serve for customer problems, we can help them modernize their infrastructure in a very flexible way. We can help them adopt hybrid cloud architectures for them to run their applications and data. We can help them build and run modern applications, which they're doing plenty of. And we can help them in this world of enterprise AI, where they're starting to build and run more and more of these enterprise AI applications that rely on inferencing from a variety of models, both frontier models and open-weight models. So that portfolio is now much broader, and it allows us to address a much broader need. And while that's happening, of course, we've also seen the supply chain crunch happening-
and prices on the hardware side going up, which of course, indirectly puts a crunch on many of our customers. But we're able to offset some of that with the fact that we have a much broader portfolio with some of these newer elements in our portfolio, like external storage ramping up.
Okay, that's a great overview. Rukmini, maybe we can bring that into the fiscal 2027 outlook. You're entering the year with strong bookings momentum, several new growth drivers beginning to contribute. But then, as Rajiv mentioned, you've got a more challenging hardware environment, slower growth in the renewal pool. How did all those factors come together in the initial guide? And what would need to develop for the business momentum to prove stronger than that initial starting point?
Yeah. I will maybe touch on three things. First, as Rajiv has already alluded to here, we have really robust and strong growth drivers for the business, including growth in our core hybrid cloud platform, including external storage. All of the cloud native and AI capabilities that we are driving, and our growing partner ecosystem. You put all those together, several core growth drivers. That is the first point. Second, as Rajiv said also, was that the supply chain environment for our customers means that if they need to procure a server on which to run our software, that has become more expensive and is taking longer. What we are assuming for our fiscal year 2027 is that those dynamics continue. We only saw it for a portion of 2026, as you will recall, Matt. For 2027, that is a headwind.
We are assuming, for example, that one of the assumptions around that we think about is orders that come in with start dates in the future. We are assuming that that percent is higher in 2027 compared to 2026. The third piece is we have talked about our renewals cohort. The first two pieces I talked about are more about incremental business, so land and expanding, landing new customers and expanding with our existing customers. The third piece, when we look at our renewals cohort, we expect that to grow in 2027, but at a slower rate compared to 2026. You put all of that in. We believe we have a lot of great tailwinds in the business that we both talked about. The supply chain is a bit of a factor outside of our control. That is a headwind.
We are trying to mitigate those as well with our support for external storage and our cloud offering that can run on bare metal. That helps mitigate, but it does not fully offset. We expect that to be a headwind in 2027, Matt, but the core demand for the business continues to be strong.
Okay, so you touched on future start dates a little bit, but let us just go a little deeper there. I think some of the business pushed out of fiscal year 2026 should be contributing this year.
but the new bookings could continue to shift to the right. So how do you think about the net effect of that and the guide? How could that change if hardware availability actually improves?
Yeah. So the net effect of that is a net headwind in 2027. Just to be clear on that, Matt, and the reason for that is we really saw some of these supply chain headwinds only in the second half of 2026.
Whereas we expect a full year of that in 2027. So it is a net headwind. To your point, what could cause us to do better than what we expect at this time, is if somehow this supply dynamic eases, which again, at this point, we don't expect that. But if that were to happen, that would help. Or another vector, for example, is our support for external storage offerings, which are to be clear, small today, but Rajiv said on the earnings call, for example, that we had seen a significant uptick into our fiscal Q4, our July quarter. So if that uptick and interest on that offering from our customers were to be much higher than we have expected, that would help as well. So those are some of the factors that could lead to us doing better than we believe here.
Broader adoption of hybrid cloud as well as the public cloud. That would be the other option.
Yeah. Rajiv, maybe staying with the supply chain backdrop for a moment. Given that hardware availability remains uneven and server prices are putting additional pressure on customer budgets, how are those dynamics actually surfacing in the pipeline today, and how do you see customers adapting their plans?
Yeah, I think you mentioned some of this, Rukmini, already, but clearly, hardware prices are now 4 to 5 times what they were a year ago, for servers. And that's certainly putting a crunch for customers as they have to manage their budgets. Now, in a lot of cases, the applications that we play in are mission-critical applications. And when it's a mission-critical application, if the hardware is end of support or whatever, you have to refresh it regardless of the price, because it's business-critical application that you're running. So in that scenario, they have to go ahead and do the transformation. There are other cases where they can defer things. If you're a public sector, for example, and you have no budget left, this is all you have for the year, and you have to go out and get new appropriations, there's not much you can do.
We have seen projects getting postponed as a result of some of these things until budget becomes available. We have seen those kinds of dynamics and, again, we are able to offset some of that by saying, in a lot of cases, we can help you modernize without you having to buy any new hardware whatsoever. You can use the servers and storage you've got, or you can use the public cloud where there's hardware easily available and use that as offsetting factors.
I guess I'm kind of curious because if you go to the fiscal 2026 timeframe, it felt like from a customer behavior standpoint, with the supply chain backdrop becoming increasingly volatile, it felt like it was very reactive in nature in terms of pushing out future start dates and requesting separate deal terms from Nutanix. When you think about fiscal 2027, given that we're now call it 12, 18 months-
into this sort of challenging procurement environment, do you anticipate maybe a little bit more steadiness in terms of the activity across the customer base? Talk to us about what you're hearing from customers themselves.
Yeah. I think, for example, future start dates will continue because that's a means for customers to anticipate the fact that we can close a deal with them, but their hardware's going to be available whenever it becomes available and so forth, and they can time their licenses to start then. That's a natural thing that we expect it to continue this year for sure. Now, I think customers are getting much more adept themselves at understanding and navigating through the supply chain. It used to be that when it came to hardware, it was like, okay, they had their set of one or two vendors that they would just have frame contracts with and just procure.
Now they're being increasingly going and checking to see what's available across multiple vendors and trying to bring in new vendors if they think they can get better prices or lead times as needed. So they're starting to adapt quite a bit. We had, for example, a large Fortune 500 customer that we talked about who decided to migrate to the public cloud, right? Using our platform so that they could keep all the applications exactly the same way they are, but that was their VMware migration, and they're actually using our database management also to run their databases on top of the public cloud using our platform. So they're starting to become more savvy about using other options, right? External storage, multiple vendors, moving to the cloud, all of these as options to help them navigate through this.
Okay, great. Now I want to talk a little bit more about sort of your diversification away from hardware. Let's start with external storage. As Rukmini mentioned, bookings increased sharply in the fourth quarter. You had several seven-figure wins, even with NetApp in limited availability. I guess, what are those early results telling you about the breadth of the opportunity and how much could that opportunity expand as those integrations roll out more broadly?
Look, I mean, we've been driving HCI for the last 15 years as a company, and HCI today represents about 20% of the install base. 80% of the install base out there is still external storage connected to third-party servers. What we're able to do now is to go after that install base and not have to convert it with an architectural shift with new hardware to HCI, but instead tell those customers that they can deploy Nutanix software in those environments without changing anything. That has a huge effect in terms of, rather than trying to fight to convert those customers over, we can meet them where they are today, especially in a world where hardware prices are high.
They want to continue to get as much as they can out of their hardware investments, and at the same time make it much easier for them to migrate to a Nutanix platform. It's a pretty significant thing. Increasingly now, we think of our platform as a much broader platform where storage is but one option. You can actually consume our platform with or without our own built-in storage, and the rest of the stack is still available. You've got computer virtualization, you've got network, you've got cloud management, you've got a Kubernetes stack, you've got databases management, you've got public cloud, and you've got AI. There's a whole stack that we can sell and offerings we can be very relevant to the customers, with storage just becoming increasingly an option for the customer based on where they're at.
Yeah. Maybe just a follow-up there. You've expressed that you've been at this for 15 years. HCI's about 20% penetrated. I guess question one is, why has it taken so long to really capture more of the three-tier architecture? Then two, do you see products like external storage accelerating that penetration for Nutanix?
Oh, yeah. I mean, I think that the answer to the second question absolutely is yes. There's no doubt, right? I mean, I see external storage support as one of our biggest growth drivers, as the biggest growth driver, I would say, for fiscal year 2027 and beyond for the next few years. And our portion of the business that's coming from external storage will continue to grow over time because we're not fighting the tide right now. With HCI, as they say in IT, there's people, process, and technology. Technology was the least of it, actually, because from a technology perspective, the fact that you can consolidate computer storage, all of this together into a hyper-converged offering, has delivered 30%-40% TCO. That's not changed. But what's required for that to happen is for customers to adopt that architectural mind share. There's different teams managing servers and storage.
They have to consolidate those teams together into one team, change their operating processes, and they need fewer people to run this infrastructure because of the level of automation. Those are people and process changes, and those things happen slowly in the enterprise. And that's why I think it's taken us this long and, again, I think like with every new technology, we started out for, let's call it, simpler applications like virtual desktops, and then gradually now migrated to the heart of the enterprise business-critical applications. So HCI will continue to grow over time. There's no doubt, and it will continue to eat into the overall market. But now for us, rather than fight the rest of the market, we can actually embrace the rest of it with external storage.
Okay, great. And then with NC2, this is another area that gained momentum in the fourth quarter, and customers appear to be using it in several ways, right? One is a bridge while they wait for hardware as part of a VMware migration or as a more permanent public cloud deployment. Which of these use cases is proving most repeatable, and what does that tell you about the longer-term role of NC2 in the portfolio?
I think what's important, I think, is we've been with NC2, we've had it in the market for several years, but I feel like its time has really come. We're now across all the major cloud providers. We have the supply chain constraints in place that make it harder for customers to get hardware on-prem. And we have cloud migration, which has been an ongoing theme for many years at this point. So we take all of those things, the NC2 use cases are plentiful. The thing that I want to say is regardless of what the use case is, there's a very high retention rate for customers. Once they use NC2 in the public cloud, they tend to stick with it. Our retention rate, we haven't talked about specific numbers, but those retention rates are very high. Once they start consuming, they continue consuming.
The use case, to go back to your first part, there is just cloud migration. When I say cloud migration, it's migration and ongoing running, right? They don't migrate and throw away the platform. They migrate and run on this platform. Why? Because they can run the application without making any changes. It performs beautifully. The application owners don't even know it. In fact, we had, for example, State Street Bank, talk about how they are using us at pretty massive scale across both AWS and Azure today, right? They actually said that their users actually got better performance. They didn't even know this happened, but they were getting better. The outcome from a user perspective was much better.
That's the kind of value that we can deliver for these customers, and they are operating in a permanent mode across our two clouds, two public clouds, and their on-prem data centers. We see it being used for temporary capacity expansion, whether it be while they're waiting for hardware or just elasticity. We see it being used for disaster recovery. We see it being used for global expansion of footprint without having to own data centers. All of these use cases, I think, are sticking, and we have customers using every one of these today.
Okay, great. Let's turn to VMware for a moment. I think some customers moved quickly following the Broadcom acquisition. Others renewed and used that time to assess their options. As that second group comes back to that decision, what's changing in the seriousness of those evaluations and the scope of the workloads under consideration?
Yeah, I think it's becoming more and more clear that in the long term, this is not the platform that they're going to bet on. In fact, if you talk to Gartner, they say only 7% of customers will remain long-term with VMware, at least from a sentiment perspective. So many of them are looking to migrate over a period of time. What we've done is to make those migrations even easier, right? The migrations are always automated, but now you can do so without changing out hardware as well. In many cases now, we're able to go in and meet the customer where they're at with their existing hardware and do a seamless migration and get those migrations done. We've done many successful migrations already. But to your point, we always said this would be a five, 10-year journey.
Customers are going to migrate in waves, depending on their situation. We still have many customers just starting their migration now and many still to start. This opportunity will continue for many years to come. It will be in phases. It depends on the customer sentiment, their timeline for renewals, their timeline for hardware refresh cycles, and then, of course, their own priorities on this versus other things that they have on their mind.
Rukmini, anything you would like to add on sort of the quality and health of the pipeline around the VMware opportunity?
Yeah, look, I think as Rajiv has said, and I think we always said it is a multi-year journey. It continues to be healthy. I think increasingly, it is hard to sort of say how much of it is because of the Broadcom acquisition versus just we have competed with them for a long time, as we have said. This, we expect, will continue to be one of our core growth drivers for the years to come.
Okay. Rajiv, maybe turning to the competitive landscape. Have you seen any meaningful change over the past year in alternatives customers are considering or the factors driving their decisions? How is Nutanix positioned to evolve relative to those options?
Yeah. Look, I think we spend a lot of time talking about VMware competes for sure, and VMware migrations. But almost the first question they say is, Okay, we might get called in for one of those, but then top of mind for all these customers is also, Can you help us with AI? That is where we get into what we can do with AI. I am sure we will cover that later, Matt. But with respect to the competitive picture, they are looking for a platform that is not just a platform that they use for migrating away from their current vendor, but also a platform that can take them into the future. Into the future when it comes to be a platform for AI, for modern applications, for hybrid cloud. From that perspective, we have not seen a huge amount of change in our competitive landscape.
The public clouds are both our partners and our competitors when it comes to landing workloads. Red Hat is a competitor. We see them. We see Microsoft occasionally on-prem as a competitor, but not very frequently. Of course, the biggest competitor we have is inertia on the part of the customer.
They say, Let us just stick with the status quo.
Yeah. Okay. Rukmini, this one for you. Beyond the core infrastructure opportunity, the fiscal 2029 framework for mid to high teens growth, I think it assumes a larger contribution from products like Kubernetes, database automation, cloud management, AI. What needs to happen for portfolio attached to reach that next level, and where are you seeing the strongest early evidence that that model is beginning to work?
Yeah, I think one thing we said at Investor Day, which I think is what you're alluding to, Matt, is that if you look at two fiscal years ago, our proportion of land and expand bookings that was coming from those solutions was a low 20s. We're saying in fiscal year 2029, that will grow to be about a third. That is a big meaningful change and means it's growing faster than the rest of the business. In terms of what it would take, I think there are some areas where it's a much more natural motion for our sellers to go out there and make the case that Nutanix, as a platform company, should be more embedded in their system. Then in others, we're investing in areas like specialist sellers.
Technical specialists who will come in once the seller's identified a deal and it's progressing to make sure that, for example, our database management platform that you alluded to, NDB, Nutanix Database Service, is one where that would be applicable. The core sellers are working on a deal, then we'd have our specialist sellers come and help them as the deal progresses and gets more technical. One thing we've done over the last few years is we have invested more
in those specialist sellers where needed, Matt, to help our core sellers drive more of that platform.
Yeah.
I think there's also market interest as we've started, we've talked about container management or NKP, Nutanix Kubernetes Platform, and Nutanix Enterprise AI as Rajiv talked about, where those markets are large and growing fast. We intend to capture our fair share of those going forward.
Just on the portfolio specialist, I know those investments have been underway for quite some time. How comfortable are you with sort of the maturity of that motion today? How successful is it in contributing to new product attached today?
Yeah. If you look at our most recent fiscal year, our year-end was July 31, we saw really great progress and great growth in many of those. NKP, the NC2 platform that Rajiv talked about, NDB, all those had a really good year, and we have high expectations for them going into 2027 as well.
Yeah.
Going forward, I think we're doing the same for AI. We're investing in AI specialists, because I think when you look at some of the motions here, the external storage versus HCI, that's something that our mainstream sellers know inside out, and they can go sell it. But some of the newer elements in the portfolio, Kubernetes, modern applications, AI, public cloud, these are where I think we augment with specialists, and that's actually proven to be working very well for us.
On the Kubernetes piece, what does it look like in terms of an RFP or a POC that you guys actually win against a Red Hat? Why typically go with a Nutanix?
Yeah. In many cases, by the way, there are situations where we win against Red Hat, and there are situations where we coexist along with Red Hat.
Okay.
Okay, also. I will give you the example from this large financial services win in Asia Pacific that we talked about this last quarter. This was a customer that believes in having two vendors for everything. For the virtualization side, they have been a long-term customer of ours and VMware. For container management and containers, they had Red Hat. They brought us in to become that second vendor for Kubernetes. That was the ticket to the large deal that we closed with them in Q4. That was a situation where we were in there as a second vendor. There are situations where the customer also wants something that is simpler, right? Red Hat OpenShift, per se, is a great platform. It is well-suited for the needs of large-scale deployment with a lot of services attached to it.
Sometimes the customer just wants something simpler, or their experience with standing up Red Hat has not been as good. In those cases, we end up displacing them, or we end up being used for much smaller cases. For example, when we look at commercial accounts, we have seen actually great success with Kubernetes in commercial accounts, especially in emerging countries like India and so forth. Why? Because there is a lot of software know-how over there. People are building their own customer applications, and those are all containerized applications. Kubernetes is rapidly becoming a mass market thing.
And so those types of customers need much simpler platforms. They don't need a very services-heavy platform. They need something that can be deployed easily, quickly, open source compliant, and that's where we win-
Okay
-in those accounts.
Okay. Makes sense. Let's shift to AI. I thought this was an interesting anecdote you provided on the earnings call. You spoke about Nutanix's experience moving more AI workloads from frontier models onto your own dedicated-
Yeah
-infrastructure as usage and costs started to increase. Is there a similar tipping point you see emerging from enterprise customers? What will determine when most of them reach it?
Yeah, absolutely. As we talk about enterprise customers, we talk about three components to what they need to think about for AI deployments. The first one is, you have agents and you have applications. They need to be running largely wherever your data is, and they will be running on CPUs. We have a platform for delivering that for them. The second is they need to consume intelligence. These agents and applications need to consume intelligence. Intelligence can be consumed in many different places, right? They can be consumed through frontier models. In that case, they just go do an API call to a frontier model, and that is it, right? Increasingly, to your point, we are seeing more and more interest in optimizing costs by deploying open weight models.
Our own internal case study was a good case in point, where we were using widespread deployment of tools for our software development effort. Because we are a software company, our engineering team has widely adopted these tools, everything starting from GitHub Copilot, to Cursor, to Claude. Our costs were shooting through the roof, especially as everything went to a token cost model. We invested $20 million in our own clusters, running open weight models, and we expect to service about 80% of our needs through those clusters-
and go to the frontier model for the remaining 20%. For that 80%, it is a one-time CapEx spend. We spent $20 million. That will go take us for about five years or so. We are no longer paying on a per token basis. We have full use of those clusters. They can be used for multiple applications and shared. It is a win for us. The ROI on that is very good for us. Within a year, we expect to recover some of those costs. That story is starting to play out. That is part two of our story for enterprises. For the ones that are using AI at reasonable scale, there are customers out there that are spending over $100 million a year on AI spend, right? They are now starting to think about how to optimize that. That is where this plays a lot, right?
Just using open weight models on their own clusters, whether it be in a colo or on a neocloud, or even on top of a hyperscaler. Makes sense. In the middle is the agent gateway or a model router that sits between the agents or the applications and the models. That's a control point for regulating access to these models, for optimizing the user models for these applications, for providing control, costs, and governance and security around these. So those are the three components that we provide. It's early days for enterprise adoption, to be clear. A lot of companies, everybody wants to get a handle on their AI spend, so the gateway piece is very critical for everybody. The ones that are starting to spend at large scale are interested very much in optimizing those costs.
As agents get broadly deployed, they need a secure platform to run those agents. That combination, I think enterprise adoption is still fairly early, but starting to take. The product market fit is there, and we have a number of wins, and that's a growing portion of our portfolio.
You mentioned AI gateways. I'm just curious because do you see this as sort of like an infrastructure first problem? Because it does seem like a lot of companies are trying to own sort of the-
Oh, yeah.
-model orchestration layer today. What's your perspective on how this shakes out longer term?
Yeah, I think there's companies coming in from different angles. We are coming in from the fact that we are a platform that applications run on. We understand the applications and what they need, and our gateway is built around open source constructs. It's built around an Envoy Proxy. That's where we come in at. Then there are security companies that come at it from a security angle, observability companies that come away from that perspective, cost governance. You can come at this from many different places, but that area, there's going to be a lot of players. Our value add is the fact that we also understand how to run applications and agents and how to provide the intelligence, and therefore that translates into how to operate the gateway for us.
But again, customers have a lot of choice in that middle layer to pick from. I think when it comes to both sides of the coin, there's not too many players. For running agents, who do you go to? You go to us, you go to Red Hat, maybe you go to VMware. Same thing on the other side. It's either a homegrown stack that the big guys can afford to go build out for intelligence, or they come to us or Red Hat or somebody like that. Those are the players on both sides of that equation.
Then maybe let's just talk about the broadening partner ecosystem, especially on the AI side, where you've got an existing relationship with NVIDIA. You announced a new partnership with AMD a couple of quarters back. You just announced something with ChronoScale. Just tell us how this fits into how you're advancing the AI infrastructure opportunity at Nutanix.
Yeah. Clearly, I think the silicon providers are driving a lot of the ecosystem. Starting with NVIDIA and now AMD coming on board. For us, it was very important to partner with both of them. So we've been working with NVIDIA for a long time. We've been working with AMD on the CPU side, but now most recently with this new partnership, we are working with them on the GPU side. The idea being to enable our full stack on top of these platforms so that enterprise customers can get access to that intelligence wherever they want. That's the rationale for these partnerships at the silicon layer. As we get into where people consume intelligence from, that's where neocloud come into play. We believe that there's multiple options. AI intelligence will be a hybrid use case.
There'll be some frontier models, there'll be some running on hyperscalers, there'll be some running on these open-weight models in neoclouds or on-prem. By working with these neocloud partners, such as ChronoScale being the first of them, we expect to be able to deliver a full stack service, for the neocloud to be able to deliver a full stack service to enterprise customers. So when we go talk to an enterprise customer, one of their challenges is, Yeah, I love the idea of optimizing my cost, but now I need to go set up my GPU clusters, and I need to go manage all of this. I may not even be able to get my hands on one. So can you help me? Then we say, Okay, why don't you work with the neocloud that already has this?
Or even with the hyperscaler, where you might be able to get access to bare metal, and we'll help you with standing up the stack and getting what you need. So that becomes a very important part of the ecosystem. We're working with neocloud to enable easy access to Nutanix Enterprise AI. Ultimately, the goal is to simplify how Nutanix Enterprise AI gets deployed and adopted, and that's really what we aim to do as a platform.
Do you expect more ChronoScale-esque partnerships? Are you able to broaden this out a bit?
We're working with a number of other neoclouds. We hope to see more of them announced shortly.
Okay, great. Rukmini, let's bring you back in. You are funding additional sales coverage and investing across external storage, cloud native, AI, and digital sovereignty, while still planning for operating margin expansion this year. How are you prioritizing those investments today, and where do you see the clearest path to further leverage of the business skills?
Yeah. I'll start with the last part of your question first, Matt. In terms of further leverage, we've talked at our Investor Day about how we think there is a path here for us to get to greater than 30% operating margins over time. For this fiscal year that we're in, that we just started, we've guided 24%-25%, which is a margin expansion from last year into this year. Our take in the near term has been to balance this need for more investment because we've talked about how we ended the year strong. We think there's lots of growth drivers for the business, and so we want to make sure we're investing in a thoughtful way to capture that growth while continuing to expand margins. That's sort of the balance we're trying to strike.
In terms of areas of investments, you touched on a few of them already. From a product perspective, we think that our innovation in several areas is really important for us to continue supporting more external storage platforms because that can only sort of help us in terms of capturing the market. Our agentic and AI capabilities that Rajiv just talked about, the sort of three-pronged approach that he outlined. All of our cloud native efforts, we think are also important areas because those are large established markets, cloud native, external storage, that are really there for the taking in terms of our ability to execute once we have the offering out there. Those are all areas that we're investing in from an innovation perspective. From a go-to-market standpoint, we are adding more salespeople.
We are hiring more salespeople this fiscal year because, again, we think that we have the capabilities, the set of products available for us to go and address this market. Rajiv said the market is really big, so we want to make sure we are capturing that, so investing in that area. Those are the two big areas I would say. Then there are smaller investments like digital sovereignty that you alluded to, that we mentioned on our earnings call, in terms of making sure that we are ready to capture the market largely outside the U.S., where they are thinking about sovereignty. We think that is a revenue opportunity that is worth addressing and going after. So we want to make sure we are making those investments as well. Generally, we think of these investments as Rajiv said, like the GPU farms that we invested in.
We will make that up in a year. So our horizon, that one was a very clear choice that we made. Some of the other ones will take a bit longer, but we still think the return is absolutely there for us to go and drive those. Anything you would add, Rajiv?
No, I think you covered it.
Yeah.
On the digital sovereignty piece, Rajiv, how developed is that opportunity today?
I would say it is a developing opportunity. There is more and more. It is, of course, entirely outside the U.S., whether you go to the European Union.
Yeah
Whether you go to the Middle East or to Asia, it's an increasing concern. There's multiple elements to that. Essentially, it's about ultimately owning and operating their own infrastructure and reducing dependence on others. What that means is first, they want to have their own clouds or their own operators operated by their citizens. They want to have ideally, data sovereignty also locally. They want to have local support. They don't want to be reliant on outside support. Then ideally, they also want to have technology independence, but that's very hard because you don't have a lot of local tech providers, so they're still dependent on U.S. technology, in many cases. But they'd like to get some insurance around that by making sure that if there's any kind of policy, geopolitical concerns, they have rights to use that for some period of time, for example.
Those are, I think, the different elements of the sovereignty thing. They're starting to be codified, in the EU especially, they are taking a lead approach to codifying some of these things, starting to be part of RFP requirements. We believe we are very well positioned to address those because we enable countries and companies in those countries to run their own clouds, fundamentally, and do so in a place where they can control where their data is provided. We are adding to our local support so that we can support them entirely in region without having to go outside.
Okay, great. Rajiv, just to step back and close out the discussion, Nutanix now has several potential growth vectors beyond traditional HCI. Which of those opportunities do you believe could have the greatest impact on the company over the next several years?
Yeah, I think if you look at it, in the long term, I do believe that Nutanix will be the platform for AI applications, because I do think AI is going to be widespread, deployed everywhere, and we aim to be the underlying platform to help enterprises adopt AI. Now, in the short term, as we look at this, VMware migration continues to be a big thing. Of all the growth drivers that we have, our ability to now address the 80% of the install base out there, that is just external storage connected to servers, is going to be the fastest growth driver, right? In the short term. If you look at this fiscal year, I think that's going to be our biggest growth driver.
But like I said, as you start looking out, the picture of Nutanix being not just a platform for today's applications, but also for the modern applications that are coming and the AI-enabled applications. That's really, I think, why this is a long-term growth play.
Okay, excellent. Well, that's a great place to leave it. Rajiv, Rukmini, thank you so much for joining us today.
Thank you, Matt.
Thank you for having us, Matt. Thank you.
Thank you, guys.
Thank you.