Hewlett Packard Enterprise Company (HPE) Earnings Call Transcript & Summary

Hi. Good morning, or good afternoon to all of you. We are delighted to host HPE management today following the security analyst meeting hosted last week. On the call today, we have CEO, Antonio Neri; CFO, Tarek Robbiati, both of whom you're very familiar with and already know. There's a lot of change that Antonio and Tarek have initiated at HPE from focus to go to market, cost optimization programs, all of which translated into the longer-term outlook that they just shared at SAM. Also on the call, we have Andy Simanek from IR, who's going to go through some disclosures really quickly. Andy?

Perfect. Thanks, Wamsi. So just before we get started, I want to take a quick moment to read the disclosures. You will hear some forward-looking statements in today's discussion. These are based on risks and assumptions that are described in our annual report on Form 10-K and quarterly report Form 10-Q. Our actual results could differ materially, and we assume no obligation to update these. More details can be found on our website at investors.hpe.com and our recent Q3 earnings announcement press release dated August 25. So with that, let me turn it back to you, Wamsi.

Yes. Thanks a lot, Andy. So let me kick it off first. Antonio, you highlighted the importance and, frankly, the strategic value of moving to as a service. And at SAM, you highlighted what the opportunity size was. You said there was $6 billion that -- that's there today and can grow to $22 billion in fiscal '23. So maybe to kick it off, can you address like what part of that $6 billion you have today and where you expect of that $22 billion, you will go to by fiscal '23?

Well, first of all, good afternoon, good evening, everyone. Thank you for having the opportunity for Tarek and I to speak to you today, Wamsi. So first of all, let's talk about what we see in the market. Obviously, we see a significant shift to consumption-driven business models, where people pay for what they need as they go. And obviously, we have seen that journey started some time ago, and that's why we have seen the growth in the public cloud. However, as we said, cloud is not a destination, it's an experience, and we want to bring the cloud to all the apps and data for our customers. Whatever the lift, whether it's off premises, on their premises, in a colo, on data center, all the way to the edge, which ultimately is where we see a lot of the applications and data move into. In that context, as I think about that, infrastructure, software and services market, we see that transition happening really fast, and that's what we talked, Wamsi, about $6 billion in 2020 growing at the 58% CAGR in 2022 -- '23. And in that, today, just to give a sense of the GreenLake business, which is the cloud that we bring to you in a consumption-driven model, already has more than $4 billion in total contract value, almost 1,000 customers, and thus generate a revenue -- annual revenue of around $1 billion. But what is the recurring part of that is what we call annualized revenue run rate, and we are already in the $500 million and above. And we see that continue to accelerate because as our bookings with our cloud services continue to grow high double digits, and we committed ourselves to grow 30% to 40% on a compound annual growth rate, reminding the audience here that in Q3, we delivered 82% growth year-over-year. That AIR will continue to accelerate. And that's why we are confident that we have the right to play and win in a consumption-driven market because in the end, it's about cost and experience, and that's what we are driving with our GreenLake cloud services.

No, that's helpful, Antonio. So maybe to put it in context for our audience, can you talk about who you view as your main competition here in this as a service on-premise world?

Yes. Well, first, we should define that the market is hybrid. As you saw in my presentation, we really know that 90-plus percent of our enterprises are already using a hybrid model. They have apps and data in the public cloud. They have apps and data in their data center in a colo. And now we see new cloud-native workloads happening at the edge because we see the explosion of connectivity and devices that ultimately everything computes. So in that context, we see competition across the board. On one hand, one spectrum, you have the public cloud, obviously, that they're trying to come into the on-premises and all the way to the edge, and you see competitors like AWS with Outpost or even Azure with the Arc or Fiji solution and then, obviously, we, on-premises with our GreenLake and then moving to the edge. And then obviously, the traditional OMS. The fact of the matter is that we think we have something unique and differentiated. And I have to tell you, in my conversation with the customers and partners, we see tremendous momentum with GreenLake and our competitors, traditional like Dell, Cisco, Lenovo, you name it, a little bit scrambling trying to get an answer to that specific GreenLake answer -- question to the on-premises and at the edge. So we see both spectrum. But remember, GreenLake is a true hybrid solution that brings the cloud to you for all your apps and data, including the public cloud because many customers don't want to be longer in the IT side -- in IT run side. And therefore, our GreenLake managed services can curate the workloads in the public cloud, deploy the cloud experience on-prem and be able to manage with one consistent experience and consumption-driven business model.

So which of your -- in your customer conversations, what sort of workloads, what sort of customers are the ones who are the first and most eager to sort of move to this as-a-service model?

Well, everything that has big data and large data sets, where gravity is a key component of that, Wamsi, is -- those are the typical workloads, where basically, the way I explain it is very simple. It is cheaper to move the cloud to the data, not the data where the cloud is. And it's no longer the cost of compute, cloud computing, the benefit of it. It's the fact that you don't need to pay for aggressive data back and forth, and that's the benefit with GreenLake. We could still leave it where it is. We can modernize it, cloudify it, automate it, with the things that the cloud experience brings to you, but we save a lot of money by not moving data back and forth, which is the biggest part of the bill that now customers have seen it. Then obviously, customer says "Well, great, Antonio. That's great. But I don't want to be even anymore in the run part." So then we said, "Don't worry. We can run it for you in your own premises, and we can automate everything with you and deploy what we call AI ops, which is basically full automation with intelligence built on it, so we can eliminate OpEx out of the equation on top of that." So big data-intensive workloads, that's where we see a little bit of repatriation what people think about. SAP is a great example, and we just announced a unique partnership with SAP. Think about new cloud, native workloads like Splunk is a great example, we used at the SAM event with Wells Fargo. I do think that's data intensive. It has perfect, what I call, opportunity to stay on-prem. And then, obviously, you still need to provide an environment for developers to do their job. And much of those works, much of those workloads need to stay access to the data, and that's why the colocation of applications and data goes together. And last but not least, to think about AI, machine learning, big data simulation and modeling, where HPC workloads are perfect examples of it.

No. That's great. So you mentioned some metrics around recurring revenue. If you look at sort of the projected CAGR of what you're talking about, it's -- in terms of revenue over the next 3 years, it's still going to be somewhat small piece of your aggregate company revenue. Do you see a need to accelerate this? And if yes, how do you go about doing that?

Well, what we committed, and Tarek can comment on this, is the fact that we're going to triple that revenue. And -- which is very aspirational and bold in many ways, but we feel pretty good about it. We actually believe we can do this because we have the momentum and the technology. The GreenLake business, first and foremost, is a software business. I want everybody to understand. It takes a lot of software, a lot of services capabilities and business innovation to go do it. Obviously, it's -- as I think about that transition to consumption, we have an opportunity not only to capture our current installed base moving from CapEx to OpEx, but capture the transition of other vendors, think about Dell, Lenovo that also will need transition to a consumption model because we have a better experience. So it's a share gain on that opportunity and then, over time, drive usage because the usage aspect of this is where we believe we have 20% to 25% incremental upside. So there is us as installed base in the market. There is the capture of the share in the transition to OpEx on-premises, and then there is the usage that comes on top of that. Obviously, we continue to see what is possible in terms of accelerating this. But in the end, remember, this is just a deferred revenue. But ultimately, our message to the shareholders, we are transforming every aspect of the business to become more consumption driven. I don't know, Tarek, if you have any comments.

No. I'd simply say the -- this is a -- the as-a-service business is, as Antonio said, software driven. And what we're doing right now is that we are winding the machine to deliver significant revenue momentum, right? Already, you're seeing $1 billion of revenue materializing in our P&L this year. But as we said at the beginning, the total contracted value is $4 billion. That total contracted value will unwind in the P&L over time as we deliver these services to our customers. We keep adding customers, and this is why this business is so strategic for us because it builds substantial revenue momentum with, as Antonio said, software margins, right? That's the key here is to understand that this is a momentum enhancer of revenue and a margin enhancer for the company. That's why we're so excited about it. And this work that we have done on the as-a-service on-prem market is very, very interesting. I'd like to invite everybody on the call to look at the as-a-service world. It's not just a cloud panacea. It's a really -- there's a vibrant on-prem world, as Antonio has described, that we can capitalize on, and that opportunity is substantial, again, in revenue momentum terms and margin terms.

So I was wondering, just with respect to GreenLake specifically, is there a M&A part of the strategy that pertains very specifically to GreenLake?

Well, listen, I mean, our M&A strategy has been consistent, right, based on very strong, disciplined return on invested capital. And when we think about assets, I think about us is that accelerate the strategy. To your point, Wamsi, GreenLake obviously is the tip of the spear and then by bringing talent and intellectual property that we can scale through our go-to market. But in the end, when you think about GreenLake, I think about GreenLake in the context of software. Remember, there is a run time and there is a fleet management that goes with it. And we introduced our HPE Ezmeral platform, that's the run time. And then we have GreenLake Central, plus every aspect underneath that. That's the fleet management in a truly cloud operating model. So we continue to look at all assets from the software perspective we can add to continue to bring that experience to the next level, to automate everything and, actually, to extract the value of the data much faster. One of the key differentiation HPE Ezmeral brings to the table. Number one is open. There's an unlock, any customers in. Number two, it helps customer reduce CapEx because you can run both stateless and stateful applications, meaning cloud-native and noncloud-native applications in the same stack. That's a significant CapEx reduction for customers on-premises. Third is the fact that we automate everything with our previous acquisition of BlueData storage and through Nimble, which we had an asset called HP InfoSight, which actually is the AIOps. Actually, we can predict 86% of the problems before it happen, and we can reduce significant amount of OpEx, which, by the way, is a margin enhancement for us, to Tarek's point, because our Pointnext services team can automate every task without having human intervention. And then third, which is probably, in my view, the most important part in the end is the ability to run big data-intensive workloads at massive scale, much faster than ever before. And the example of Wells Fargo, we actually proved that the specific Splunk workload, we're able to perform 17x faster than the VMware or the Red Hat OpenShift. So we are extending the ML Ops out to this to basically establish ourselves as the choice in the marketplace because that data gravity is on-prem. And if we can bring the right capability there, we can accelerate. But in order to accelerate that, we will continue to invest in more software as we go along the way.

Okay. Great. If we could switch gears just a little bit to talk about the near term. I mean, it's been obviously a very challenging environment, not to say the least. Tarek, you reiterated the baseline starting point for fiscal '20. Can you maybe talk about what is shaping out differently relative to expectations as we started this quarter?

Do you mean in revenue terms or anything specific? What -- could you just clarify your question, please, Wamsi?

Yes, Tarek, just as you sort of spoke about the backlog a couple of quarters ago, and you made some progress on that last quarter, wondering if that's still maintaining a similar trajectory or if anything has changed with respect to that.

Yes. No, I'm glad you asked the question. That's why I tried to figure out whether you were after the backlog issue. Look, backlog, it's an issue that materialized in fiscal year '20 and is behind us. We are very comfortable to have resolved any backlog issues at this point. We did solve $500 million of the backlog issue we had in Q3. We said there would be $250 million to still come through in Q4 or FY '20. We have made tremendous progress towards that, and I'm comfortable in saying to you that by the end of fiscal year '20, it's a complete reset. We're not entering fiscal year '21 with an augmented backlog or backlog issues just like we experienced, which is a function of the disruption in the business that were caused by the pandemic for the most part. So this is going well. And so then I want to take the opportunity here to say to you and the audience, and when you look at our performance and you normalize for backlog, it's really essentially a shift of revenue on a quarterly basis during fiscal year '20, no more, no less. But when you start to strip out revenue from Q3, remember that $500 million is made out of orders that came from not just Q3, but Q1 and Q2, even if you were to strip this out and you were to strip out the reduction in backlog we will show at the end of Q4, this is the best way to look at what's happening to the underlying demand. And the underlying demand from our standpoint is steady, and we feel very comfortable about this. And this is why this has prompted us not only to give you all guidance for Q4, but also to give you full year guidance for fiscal year '21. Antonio, would you like to add to that?

Yes. I'd just say, Wamsi, number one, we are, I'm looking at my calendar, 10 days away from the end of the quarter. Obviously, we cannot share more than what Tarek said. But I will say the backlog is no longer an issue. It's gone. We have done a remarkable job in just 4, 5 months recovering from a very big disruption. We have made our supply chain even more resilient than before. And just to give a sense, yesterday, I was with partners on the phone. We had probably 50 of our largest partners on the phone and I'm prompted. One of my partners told me, "Antonio, we listen to your strategy. We love strategy of going forward. We have to come with you because, obviously, 70% of the business go with -- through the channel partners." And he said, "I just want to give you a feel, but I thank you that you guys, your supply chain was stellar in the last 4 months. And by the way, now, it's a point of differentiation for Hewlett Packard Enterprise, and we see some challenges with some of your competitors." So point is, in 10 days from now, when we close the quarter, that backlog is out of the table. It's all about new demand, new orders. And to Tarek's point, we see steady. And when we see the demand is a little bit shifting in different areas, too, because we see IT resiliency, AI, machine learning, remote connectivity, virtual desktop solutions, anything that accelerate that digital transformation and becoming more digitally native as a company, meaning the enterprises, is a necessity, as I said in my comments during the security analyst meeting. Digital transformation is no longer a strategic priority. It's a strategic imperative.

No. We appreciate the view on that. I guess, is there any color, Antonio, that you can share on what's happening across regions? I mean, it feels as though that there is some risk of COVID resurgence again in certain areas. So any color that you can share around sort of what you're seeing across regions?

Well, let's start with ourselves. We, in the United States, as a company, we're still closed in terms of the offices. Our factory, obviously, is operational because we serve a unique set of customers that need to be made in U.S.A. Outside the United States, most of our offices are open, but they are, what I call, between Phase 1 and Phase 2, phase 1 being up to 20% of the employees. There is no question, Wamsi, that there is an uptick on cases. Think about the core went up. That went down, and now it's gone up. The question is, will it be as high as the other peak. I'm not sure about it because, obviously, we have been now 6, 7 months operating. In this environment, people are more conscious about things and how they do things. And obviously, there's more mature in the system. However, we enter in the winter in North Hemisphere, and that's going to shift. In fact, my mom still live in Argentina, and they went through the brunt of it. And now they're coming out of it slowly. But boy, definitely, there is resurgence. In that, I will say the United States is very steady. I will say we feel pretty good about there and their pockets. So Europe has continued to be normal. But it's going to be a little bit of instability as we go through. But I think I will wrap all of this up by saying the demand is steady across the world. I think Asia is becoming stronger. In my view, Asia is becoming now stronger. They struggle last calendar quarter, except China. With China, they recover quick and they move down. But the rest of Asia feels to me, in this particular core, they are stronger. U.S. steady. And then Europe, you have some strong pockets and other ones that they're still managing through as we speak.

Okay. That's great, Antonio. You mentioned China, and obviously, you have a very unique segue into that market with H3C. There have been a lot of investor questions on H3C. Particularly, what are your thoughts on exercising the put option with respect to H3C over the last 2 years? It seems like the relative percentage of value that's coming from H3C relative to the total market cap of HPE has just been growing over time quite significantly and very well. So just wondering what your latest thoughts are around this. And what would be the factors that you look at to determine that you would like to exercise the put option at some point?

Well, I will start, and obviously, I want Tarek to comment on this. Listen, we are incredibly pleased with the relationship we have with a partner there, which is the unit Splendor group. We are incredibly pleased with the performance of the new H3C group. Let's remind ourselves, China is still the largest -- the second largest IT market on the planet. And we get to participate a unique setup, right? And we knew that 6 years ago when we started this whole process, we want to participate in the second largest market with an asset that will decay over time or we want set ourselves in a different way that we can have an asset that can grow and yet collect the dividends associated with that, while we also resell to them our own products. As the put became available to exercise, which was May 2019, we have 3 years to exercise that put. So we're still almost 2 years out. It's 1.5 year out, we continue to assess what is the best return for our shareholders if you exercise the put? Because on one hand, you have, to your point, right, there's a lot of dividends and an important component, which, by the way, we don't get credit for it. So let's start with that. Maybe we should get credit for it, and then we can decide what to do with it. And then -- but reality is that if I exercise the put, I have to create equal or even better value. Obviously, we want to create even more value. And to me is what areas we want to potentially invest that we exercise the put. And that's why we continue to assess what is the best return for our shareholders in the context of our strategy. But will enteric to comment because what you see on that point is that as the HTC performance continues to improve, the put value at the core multiples continue to go up and up. So we got to pick a time, and at the same time, we also have to pick what we want to do with it and obviously, work with our partner there. But right now, it feels we are in a good position as we speak. Tarek?

So Tarek, thanks for that. There have been a lot of questions around sort of the buyback in this context, right? So when we think about where the stock is and as Antonio said, you're not really getting much credit here for that put option, frankly, like the aggregate companies multiples are fairly depressed. Why does it not make sense to exercise the put option, have that capital to buy back a lot more stock or alternatively, use that for incremental M&A? I mean you guys have been very clear about driving growth in the business as well. So can you maybe address those in terms of just capital allocation priorities?

Yes. Thank you, Antonio. I would simply add to what Antonio said that, so far, in the first 9 months of the year, as a reminder for everybody, the equity interest that we have received from our stake in H3C has grown 20% year-over-year. And we are in a unique position across any industry sector with the setup we have to really create and realize value from China, which is truly remarkable. Now the factors question that you asked, Wamsi, is -- the answer to the question was what would be the factors that would determine our desire to exercise the put, fairly straightforward. As long as that equity interest continues to grow at a faster pace than our cost of capital, we should let it accrete. And we have time to make the decision, as Antonio said, until April 2022. And we're very pleased with that setup, and we feel we're in a very, very good position. No rush. We have a locked up exercise multiple of 15x trailing 12-month earnings. So as the earnings accrete, the value of our stake increases. And we will have the time to make the decision around optimizing that value for our shareholders.

There have been a lot of questions around sort of the buyback in this context, right? So when we think about where the stock is, and as Antonio said, you're not really getting much credit here for that put option, frankly, like the aggregate companies, multiples are fairly depressed, why does it not make sense to exercise the put option, have that capital to buy back a lot more stock or, alternatively, use that for incremental M&A? I mean, you guys have been very clear about driving growth in the business as well. So can you maybe address those in terms of just capital allocation priorities?

Sure. So first and foremost, our capital allocation framework prioritizes 2 things and at the same level. One is investment in the business to fuel the growth. That's really critical given where we stand. And the second one, at the par with the first one, is return to our shareholders by way of dividends. This is really key, and we have a track record of growing and paying dividends to our shareholders. And we are determined to maintain that track record as much as we possibly can. You know also that we have embarked on a strategy of fundamentally transforming our core and transforming our core businesses, particularly compute and storage, but the entire company is going through that transformation. That transformation has a cost, and that cost of transformation under our cost optimization and prioritization program is $700 million in fiscal year '21. And that depletes our free cash flow guidance in fiscal year '21 to between $900 million and $1.1 billion for fiscal year '21. So if you were to add back the $700 million, you could see, first of all, that the business is very capable of generating high free cash flow. But also when you really look at the $1.1 billion at the top end of the range, you would also have to consider that the dividend payments we currently make represent about $154 million per quarter. So we are returning effectively through the dividends something like $600 plus million per year, which is right in the middle of what we said in terms of our capital return policy, which is about 50% to 70% of free cash flow to shareholders, right? So it's right, consistent with what we said in the past. Now buybacks are part of the consideration, of course. And if we are in a position where we generate excess cash, having funded our investments and our dividend payments to our shareholders, we may take an opportunity to execute a buyback in that context. But I want to remind everybody that between fiscal year '17 and fiscal year '19, we executed nearly a $7 billion capital program, of which $2.6 billion came from dividend. The rest came from buybacks. And this hasn't truly helped our multiple, right? So what we have to consider here is the priorities in our capital allocation framework, which are to drive the growth through investments, return cash to our shareholders by way of dividends and, if and when we are in a position of excess free cash flow, execute buybacks opportunistically as we see fit.

Okay. That's super helpful. Maybe just as a quick segue. I mean, at SAM, you gave a trajectory around free cash flow, and you highlighted sort of how the free cash flow can get back to sort of a $2 billion-plus level by fiscal '23. Clearly, there's a lot of work to do to achieve that. And you guys have embarked on a lot of different cost actions and other initiatives, including what you've done inorganically with Cray and some of the other assets. So can you maybe talk, Tarek, about what are the things that are going to cause the puts and takes to that free cash flow dynamic of getting back to $2-plus billion? What are some of the heavy lifting that needs to be done to get there? And how can investors have a high degree of confidence that you will get there?

Yes. Thank you for the question, Wamsi. Let me start by saying, in fiscal year '19, we generated in excess of $1.7 billion of free cash flow, and this after paying, I will now forget this, more than $660 million on a settlement with DXC, right? So the cash flow generating capacity of this business is well north of the $2 billion, right? Let's not forget that. Now in this fiscal year, and because of the disruption caused by the pandemic, which wasn't a foreseeable event, as, of course, you totally understand, we have to rightsize the business given a new shape of the recovery, which we always said will be gradual and U-shaped, right? And it's -- we've been proven right on this one, and we went out of the gates with our cost optimization and prioritization program ahead of the market, right? And I think -- and to this day, Antonio and I are saying, that was the right decision at the time. And hindsight is always a luxury, as we all know, but at the time, with the information we had, we made the right decision, right? So this is going to cost us some money to actually reposition the business in a very strong position to accelerate out of this crisis. You've heard from Antonio that digital transformation for our customers is no longer an option. It's a strategic imperative. You've got to do it. We feel very good about that. That's why we've taken those measures to position ourselves stronger, accelerating out of this period that we have been living through. So now what do you infer from this relative to free cash flow, right? So our guide is at $900 million to $1.1 billion. If you add back the $700 million, which are one-off, you're at $1.7 billion. Then you may say, "Well, you did $2.3 billion. How could you get back to that level?" Well, I would say to you, we will, over time, as we continue to grow revenue, particularly the margin-rich revenue that we have highlighted at SAM. But also we have to go after opportunities that present themselves that are going to consume some degree of working capital. And here I want to be pointing to 2 very important opportunities for the company. One, as a Service. You heard it from Antonio. This is really, really critical. We've got to win that battle. We are probably best positioned to win the as a Service on-prem battle that is taking place as we speak. That has an element of working capital intensity that we'll have to digest. The second substantially large opportunity that we want to flag, and we have flagged at SAM, is everything that happens in the HPC, MCS and Exascale world, right? This is a world where you build very, very large systems. They're bigger than a couple of football fields or basketball fields, in some cases. And it takes time, it takes time to build these machines on-site for our customers. You have to procure all the components through our supply chain to build those machines. So from a cash flow standpoint, you have to incur the expense, which is affecting working capital. And eventually, you'll recognize the revenue over time as the machine is being delivered and accepted by our customers, right? We have, so far, won more than $2 billion worth of these systems. And there is a very clear visible pipeline of $5 billion that we can go after. We have a very strong track record of win rates. 5 out of 6 systems that were put out for bid were won by HPE, which is truly remarkable. And it would be remiss of us if we didn't go after that opportunity that exists in HPC and Exascale. I spoke a lot here. I'd like to have Antonio add more to this because...

I'm just going to -- you answer the free cash question. Now let me talk about the opportunity, Wamsi, because it's important that the others understand. Listen, we live in this new age of insights. And in order to get insight, you need a lot of computational capability. Cloud, obviously, is a way to go about it. But when you come to these very large data sets, intensive workloads, you need data and compute that could be colocated with high-speed interconnect fabric that reduces latency and can process data at the speed that the business needs. Great example of this are autonomous driving, fluid dynamics. Think about research, COVID is a great example. And what we're building here is not great just for the business, but for the country and society. The fact that we will be able to process 1 billion transactions, so it's 1 billion square or second, it's just remarkable. And just between SAM, which was Thursday, think about it, Thursday last week and today, which is Wednesday, we already won an incremental, since we told you the number, another $300 million, Finland, Australia and a very large U.S. public sector just today. And the reason why we are winning is because we are the only ones who understood this. Honestly, there is nobody else left in this particular market. You have the Japanese Fujitsu, maybe Atos in Europe. IBM clearly is no longer focused on this space. And we have the entire stack, the entire stack. Normally, we're in compute, it's over commoditized. Yes, this is a different version of compute, but we are on all the foundation technologies. What people misunderstood about Cray is the fact that we brought a company that has technologies and software to build the system. It's not another server, and that's higher-margin for us. And because you need to deliver a fully integrated experience, we are the only ones who can build these systems. And the growth is there. We highlighted -- I think Pete has highlighted the subsegment of the Exascale market. But I have to tell you, obviously, with the geopolitical situation we live, this is going to continue to accelerate, and we feel good about it. But it's capital-intensive in many ways, and that's why revenue is going to be a little bit lumpy and bumpy in this state because one quarter, we may show up and have $1.5 billion in revenue just in this business, and another one is going to come down maybe to $750 million. So you need to make sure you understand those. But boy, in the long term, this is billions of billions of dollars. And that's why when you think about servers, so we talk about giving you insights in each of our segments because we are unlocking operating leverage in each of the segments, so that we can harvest dollars to invest in the growth. When you think about compute plus HPC, that's the whole server market that IDC reports their market share, obviously. And in aggregate, we're going to grow that market because HPC is exploding. And then on the compute side, we are focusing on the profitable segments and where the growth is, particularly Tier 2, Tier 3, SaaS service provider in telco. Telco is a big opportunity because of the 5G, together with HPC and the software. That's why we feel pretty good about it.

Yes. Antonio, so congrats on the incremental $300 million deal that you just signed. What is the time frame that this $5 billion opportunity is coming up for bid? And can you just clarify for all the investors on the call if any of that $5 billion opportunity has been contemplated in your 1% to 3% growth rate that you've given from a longer-term perspective?

Well, all those opportunities are now in the pipeline, and they're all kind of in the process, call it, whatever sell stage, some in stage 2, some in stage 3. So I will say, in the next 2 years, those businesses are going to be awarded. In some cases, even '21. Then it becomes, all right, it takes time to build these things. Just to give a sense, our factory where we build the systems is a state-of-the-art factory. We have more than 6 megawatts of power because when you put the system, you have to test it for a period of time. And because a lot of these systems are water cooled, you have 1 million gallons per second moving through the systems because we are cooling the system through water. It's just an amazing marvel of engineering. And -- but the reality is that it's less about the word, Wamsi, it's going to be how long it takes to build it and then ship it. In some cases, it takes 3 to 6 months to build it. Then it may take another 6 to 8 weeks to deploy it because we have to bring it back together. We put the whole system in the factory as it was running on the customer side. We take it apart, put on track, we put it back together, and then we have to run the test and the workload. Only when the customers run the workload is when they can give us the acceptance, and that's where we can recognize revenue versus in the other traditional businesses, as soon as it shipped from the factory, revenue gets recognized.

Okay. Sorry, let me also add and clarify one thing, Wamsi, which is the $2 billion that we won, obviously, are part of our plan, our 1% to 3% long-term growth. The $5 billion opportunity or -- is obviously not. It's incremental, right? And the more we win of those deals, the better it's looking.

No, that's great. No, I appreciate you clarifying that. Maybe, Antonio, switching to compute. From -- when we look at your guidance in terms of the flat sort of ex-Tier 1 outlook for the TAM for compute. But when we look at like the rest of your segments that are all should be growing at a pretty healthy levels, so HPC, MCS, we just spoke about the huge opportunity. Aruba has got, again, like a pretty large opportunity. So the inference in that 1% to 3% is that your compute will sort of decline, core compute will decline. So, a, is that the right inference? B, what is it that can be done to change the trajectory of that or improve the growth elements there?

Yes. I think you need to separate a couple of things, right? One is compute is obviously more commoditized than any other business. You need to start with the units first and then go through revenues because, in between, you have one important factor, which is AUPs, and AUP is driven by a combination of options attached to the server and the cost associated with those options because, obviously, commodities, it will take time over time. So we believe and consider where we are. We're going to grow units in compute and then, from there, what AUPs will do. And then remember, they are cyclical. As the iPhone and the PC goes, so the rest of the commodities in the market. And right now, it seems to be that there will be some sort of small inflationary trend on the commodities, particular memory, which is one of the key components that go with the server. Think about CPU, memory and storage, right? And then there is structural pieces that comes with it. As the new technology gets deployed, particularly faster CPUs, lower nanometers and so forth, you have more memory channels that get available, which means you can attach more options. So in the same server, you have faster chip or better in many ways. That's the way I think about it. But then you can add more storage and more memory. And the unit growth, obviously, will come from the areas where we have these segments that we see growth. I'll talk about Tier 2, Tier 3 and the telco. Telco needs to deploy this virtualization of the network because, otherwise, you can deploy 5G. So there's a lot of puts and takes. And I can tell you right now, the visibility of our commodities like 3, 6 months out, then nobody knows what's happened after that. But we believe we are well positioned. At the same time, we are transforming that business to become more cloud native, and that's the SaaS that will be added on top of it because you have hardware sell, you have hardware maintenance, and then you have the software that gets attached to it to deploy these computer sources in a SaaS model that also has a component of subscription-based and a support maintenance that goes with it in that software space. So that's how we think about it. Obviously, HPC is much simpler to understand, but also lumpy. And then you think about the rest of the business, we haven't spoken yet, Wamsi, about edge. To me, the edge, particularly with Aruba, is a big opportunity for us. And through the acquisition of Silver Peak, we feel extremely bullish about that business because now we have a true complete portfolio that will allow us to deliver that edge to cloud experience from the connectivity perspective.

So one of the things, Antonio, that you've spoken about in the past is just sort of the sales motion that is changing at the company. Can you talk about where you are? What are some of the changes that have been implemented so far? And what is the results of that have been?

Yes. So let's go back, Wamsi, to HP Next. I was the architect of HP Next, and that transformation was about re-architecting the company from the grounds up. There was component about process simplification and optimization, IT and, obviously, go to market. And on the process in IT, we are well underway. By the end of 2021, we'll have deployed the vast majority of the new capabilities, which are essential to unlock operating leverage for the company because we become way more automated and simple to do business, and we remove people out of the equation. And honestly, we are much faster, much more agile to react to the market demand. On the go-to market, we have been -- think about the degrees of gray. You have a black on one side and the white on the other side. And we have been pivoting this slowly also to make sure we don't leave revenue on the table as we pivot resources. That's why the cost optimization, allocation of the resource is so important. At that time, what we did, we removed layers. We took layers out of the equation. We used to have it worldwide and a region and a subregion and a geo. Now there is no longer that. That is me and the geo. That's it. Everything else is gone. And by the way, nothing bad happened. We went in a very flat organization where the General Manager of the geo, the Managing Director and I have a direct conversation with my Global Chief Sales Officer. That's it. There is no more to do there. I mean, now what we're doing is pivoting the actual coverage in the market to the areas we want to drive growth, and we are becoming way more efficient on the traditional businesses and the way we cover it with our channels, where, basically there, we have a more distributed model where quotas get deployed in a different way. It's more transactional. A lot of this business, by the way, in compute is more like RFPs and e-auctions and you name it, and then retrain or hire new people in these new areas, particularly as a Service or the edge that we talked before, and that's a journey we have been going on. We have been very good about segmenting the market in many ways to, what I call, the RAD model, retain, acquire and develop. On the retain side, we're more efficient with inside sales, covering more of the market. The acquire is much more hunting oriented. And then develop is about growing the share of wallet, particularly using our AP&S, adviser professional services, and as well as our presales team. That's the journey we are going on. And in many ways, it's probably one of the biggest transformation we will have to complete because in the end, it's how you drive this conversation with customers. When I talk to customers, they know ProLiant, the best franchise in the server in the world. What they want to know is what ProLiant does for them in order to deliver business outcomes. We've been shifting that conversation to workload optimized solutions, are you deploying the right cloud experience and how you harvest the data insights from the data much faster. And we bring that solution to customers, and that's why this pivot is so important.

You also noted doubling down on growth opportunities. So when you think about the level of investment that needs to be made to make sure that these growth opportunities translate into the revenue growth that you're hoping to achieve, how would you characterize or maybe bookmark? Like what is the level of investment that you think is needed in these businesses over the next few years?

Well, first, let's remind ourselves, we've got to transform compute and storage, so we can free up that operating leverage, and I explained what we're doing there at the security analyst meeting. And I feel pretty good about where we are on that journey. And then take the money and reinvest in these growth areas organically and then through M&A, as you asked me early on. In the HPC space, we feel pretty good about it. I have to tell you, there is not a lot to be done there. It's more about innovating on the software side. But we have our own software. It's all cloud-native in many ways. And that's why through the organic investments, I introduced HP Apollo when I was running the compute business in 2013. Then we -- I acquired SGI. And then last year, we acquired Cray. We brought it all together. We have not only the best-in-class HPC, high-performance computing business, but also the best-in-class in-memory compute business because that's a very important aspect, especially when it comes down to mission-critical transactions, banking, telco and so forth. So there, I feel very, very good about it, but there is all about delivering against the opportunity. In the edge, obviously, you saw what we did with Silver Peak, very timely. And the COVID has proven that customers now are going to live in a massive distributor enterprise, where your home office is the micro branch, then you have branches and you have campuses and you have data centers and you have big clouds. The ability to connect all these edges in the clouds have to be software driven. And as more cloud-native application gets deployed, what that means is that the traffic gets routed to the Internet, now through the telco fixed network side. And with Silver Peak, with together with Aruba, now we have the full complete portfolio to connect all the edges on the cloud in a very autonomous driven way. And you will see more of that as we integrate Silver Peak with Aruba Central. So there, it's continued to accelerate that innovation. Security is a very important aspect of that. And then, obviously, we talk about the GreenLake and the software. But to give a sense, our plan is to -- in multiples. I'm talking about 5, 6, 7, 8x multiples investment shift into those businesses, while we transform the core business. That's the way we think about it.

Okay. That's great color, Antonio. I know we're coming up on time here. I want to ask you, one of the other changes you made recently was regarding stock comp, and we're moving to our non-GAAP and excluding stock comp from EPS. So can you help us think through what were some of the reasons to do that?

Tarek, I guess, you should take that, and I will add something at the end about talent.

Yes. So look, the practical reality is the vast majority of our comparable companies account for stock-based compensation on a GAAP-only basis. We're 1 of 2 companies who, in our peer group, didn't. And we feel that it was important to align because, quite frankly, no one ever gave us credit for having stock-based compensation on the non-GAAP side of our P&L. And so whenever the market looks at us, I didn't see many people making any sort of adjustment on that front, which was disappointing. But it is the reality, and all we're doing is realigning our reporting with what the rest of the market does, right, and that is the whole intent here. And moving forward, we've got to think about this in terms of talent, and that's also an important part of our company attractiveness for future times that we want to be acquiring. So Antonio, do you want to say something about that?

Well, that's important. I mean, listen, I mean, if you live in Silicon Valley, you understand what's going on. It's hard to attract and retain talent here. Obviously, a lot of our performance is compensation actually. It's performance-driven. We have a performance-driven compensation system. And in that, the stock is a key component of that. But the ability to be more flexible there is critical without diluting our shareholders, and that's we have been looking at this for some time. This makes complete sense. And as Tarek said, we were 1 of the only 2 companies that were not doing it that way. One hand, you have startups, right? And the other hand, you have big cloud companies. And in between, you have to figure out how to compete and win for the best talent. That's why we did it. In the end, it's just a shift from one side to the other one, but it gives us a little bit more flexibility in that context.

Would you say it also opens up maybe the opportunity to look at like a different subsegment of M&A targets that maybe you...

Absolutely, absolutely. That's one of the challenges when we do M&As is -- first is the purchase price, and then it's the retention. Then there is all the things we have to do, and it obviously comes in the way. But I have to say, Wamsi, I'm incredibly proud of the work we have done. We have done the right acquisition for the company that we are. And we have delivered accretive value to our shareholders through those acquisitions, starting with Aruba and the last one in Silver Peak. Those are a great example. And in between, being able to bring that value and through IP and talent. So -- but we've got to move faster, obviously. And definitely, this helped us in that direction.

Well, I think we're just about out of time. So maybe, Antonio, is there anything that as ending message you want to leave for the shareholders here on the call?

Sure. Well, first of all, Wamsi, thanks again for hosting us today. If you haven't watched the Security Analyst Meeting, please watch. It's available on replay on our Investor Relations website. I think we deliver the most compelling strategy that we can drive today for our company. It's aligned to the market trends. It's aligned to what we see from the customer needs. It's aligned to what the demand is going to be. But I think, in the end, we deliver a very compelling investment case. It is an investment case that takes time, right? It doesn't happen in the next quarter. But ultimately, I am more convicted than ever that we have not only the right strategy, but the right portfolio to compete and win. That's the bottom line. Thanks again, Wamsi.

No, thank you so much, Antonio. Thank you, Tarek. We really appreciate both of you taking the time and being so candid in your answers. We really appreciate it, and good luck with all the execution here. And we look forward to speaking with you soon.

Thank you.

Thank you.