HPE CEO Antonio Neri On Data Management, Teaming With VMware, And Why Competitors Are Going To Hit A Hardware Wall

Neri On HPE’s Innovation Advantage

Hewlett Packard Enterprise CEO Antonio Neri says that the company’s focus on game-changing innovation like edge computing, memory-driven computing and data management is driving big opportunities for partners.

“[The edge] is a huge opportunity,” said Neri in a question-and-answer session at The Channel Company’s 2018 Best of Breed Conference Tuesday. “About 75 percent of the data is created at the edge and it starts with connectivity, security and, ultimately, how we process data in real time. And that’s why as we think of the future of enterprise, we see an edge-centric cloud-enabled enterprise.”

HPE is squarely focused on a “partner-first” strategy as it brings to market edge-to-cloud architectures with “the right infrastructure, the right software, the right services and the right consumption” models.

“Obviously, we need partners to come along with us because customers want to work with a partner of their choice,” he said. “And this company has been built on partners like the ones in this room. This has been and will continue to be a partner-first company.”

Can you talk about the work you did with [HPE board member and former HPE CEO] Meg [Whitman] before taking the CEO job eight months ago?

If you think about the last seven years, it’s been a remarkable journey for us. And Meg drove the biggest transformation in corporate America. And the reason why we did that is because we felt that not only the digital disruption, but the pace of technology, is moving at a speed we have never seen before. Put aside all the transformations, Meg and I worked together on what is the strategy for Hewlett Packard Enterprise going forward. And at the core of that, the board felt that this is the time to bring someone in with more of a technology background.

What is the opportunity in edge computing?

[The edge] is a huge opportunity. About 75 percent of the data is created at the edge and it starts with connectivity, security and, ultimately, how we process data in real time. And that’s why as we think of the future of enterprise, we see an edge-centric cloud-enabled enterprise.

And that’s why Hewlett Packard Enterprise is really focusing on edge-to-cloud architectures with the right infrastructure, the right software, the right services and the right consumption models around that. And that’s the biggest opportunity we have as a company.

And, obviously, we need partners that come along with us because customers want to work with a partner of their choice. And this company has been built on partners like the ones in this room. And this has been and will continue to be a partner-first company. In fact, in some regions we are now reaching 90 percent of our business going to partners. I was in Asia-Pacific three weeks, four weeks ago doing reviews with our channel partners and basically the stats are very clear. Eighty-seven percent of our business goes through channel partners [there].

What do partners have to do to meet and exceed customer demands?

Obviously they need to know how to implement cloud; they need to deal with the security challenges. They need to leverage technologies like AI, and all of that has to come together in a solution. Services play a huge role, but also the consumption-based aspect of this [is important].

What are some of the biggest opportunities for partners looking at the full HPE portfolio?

Being very pragmatic, I think there’s a big opportunity to sell Aruba.

Aruba is a platform-based company. At the core, it’s all about the software and the experiences we provide through that software. I’m really proud of the Aruba team and what [Aruba founder Keerti Melkote] has done. That was my first acquisition three years ago and it has been a home run and customers love what they do because they enable the digital transformation.

If you go to our new campus when we open it in the February to March time frame, you're going to see a state-of-the-art digital experience, but it starts at the edge. And so we have quite a significant amount of investments at the edge to really drive the right connectivity with the right security and the right data processing wherever that takes place.

If you go to a health-care facility, that's a whole different experience. If you go to Vegas, that’s a different experience. So there is a tremendous opportunity for us working with you to provide those unique experiences and make the right investment to capture a huge opportunity.

Ultimately, it starts with the data and how we manage that data. Data has gravity and then I always say it's cheaper to move compute to where the data is and not to move the data to where the compute is. And there are a lot of use cases that require the compute to be closer to where the data is created either because of latency, security, compliance and, obviously, cost. And what we see today with the cloud is actually many customers, they think in their cloud strategy. They realize that there are certain things that they should own in their own cloud because it's cheaper to operate at scale and they still can get the consumption-based model. But ultimately how we provide that connectivity from the data perspective, where the data is generated, where the data is processed from an edge-to-cloud architecture. And that's where we, as a company, are uniquely positioned because we have the edge and we have the core. The core meaning traditional cloud, data centers and so forth. So I think you're going to see tremendous innovation at the edge and at the future of that is obviously building a platform, obviously cloud-enabled, AI-driven and focus on those experiences.

Talk about innovation at the core.

If you go to the core side, software is key. We are obviously a strong infrastructure company. We have the broadest portfolio when it comes down to compute.

We talk about innovation and security at the core. Think about what happened last week with people potentially penetrating infrastructure. We feel that we have an obligation to these customers to not only deliver the best compute internal performance and cost with power consumption, density and new technologies, but security. And that's why we developed with our Gen10 platform because then we can actually monitor what's happening to the server at the root level, not at the software layer level, and so that has been a phenomenal aspect for us, but as we go along—whether it’s convergence, hyper-convergence or composable which for us is basically how we bring that public cloud experience to customers—the key is software magic and the common denominator of all of that is HPE OneView.

HPE OneView allows you to deploy, provision and manage all that infrastructure with a traditional three-tier approach, what is converted to merge what we call the super set of composability, which is HPE Synergy.

What is the data management opportunity for partners?

We are in the business of data management, not storage. And you're going to see going forward a tremendous shift in the way we think about data management. And, by the way, the services piece of this is huge, so we were the first to come with consumption-based models four years ago. Now we have a new version of HPE GreenLake, which is more geared for the channel partners as well. To give you some perspective, we already have 450 customers with more than $2 billion under a global framework of consumption-based models. It started with Infrastructure as a Service. Now we are moving to outcomes as a service.

Can you talk about the changes in partner compensation aimed at driving high-value, high-growth products and services?

We see what customers are consuming every day and what’s driving growth and profits. So whether that’s SimpliVity storage or hyper-converged or composability, you get a five-to-one return, which is significant. Not only is it higher margin but it’s the architecture the customer is looking for. Because in the end what customers are going to look for is something they can turn on, give me the API to start operating my environment quickly versus going the traditional way to design and stack and rack things. Now, that doesn't mean we don't care about the volume business. If you want 10,000 servers, we want to be with you. But more and more you see what's happening. What's happened is customers don't have the time, they have less people, less budgets. Now we can actually provide the full stack that's fully optimized, roll it out and then turn it on with a few clicks and then start running the business and that's the volume-growth aspect of the portfolio. For those who still want cheap commodity hardware, we still have one of the best platforms on the planet.

What I'm excited about in the future is that as we think about a data-driven approach, we believe that two years from now we are going to generate twice the amount of data that we generated in the entire human history. So to give you some perspective: We generated 1.2 quintillion bytes a day and two years from now … it’s just astonishing what we can do. So the core architecture is not going to scale the way we need to and either it’s cost or performance or space. And that's why we have this concept of memory-driven compute, where we make the data the core of the architecture and we bring the CPU to the data in a way that’s more economical and, ultimately, we bring new technologies along the way. And that's why I'm so excited by what we're doing with the team with silicon photonics. It’s actually not too far off. That’s why we bought Plexxi because Plexxi provides the data fabric network in silicon photonics. Think about it: In a system today, you have copper running everywhere. That’s friction; it generates heat. When we go to the new design, there is no copper. It’s all about light. And the cost differential is enormous.

Are competing infrastructures going to hit a wall?

We have been talking about the fact that Moore’s Law is going to hit a wall. I mean less than 10 nanometers. Then I think we'll see seven, maybe four and then after that, what? You’re going to have it implanted in your head? It’s not going to work. And the issue is that every time we do that obviously we're trying to manage the cost of power consumption, but the reality is it’s not scaling at the same pace that data is scaling. And so that means you have to glue things together to catch where that data curves. And so we thought about how we do that and we moved the computational side above the data, not below the data. And that’s why we have to make memory and storage the core of the architecture.

Talk about the competition a little bit. You’re fighting Dell Technologies, but at the same time you’re a big VMware partner. How’s the relationship between you and VMware going?

We have what I call the pure-play competitors, you know, Dell, Lenovo and what I call China Inc. The other one is obviously the public cloud. And then you have a bunch of stops in the middle. And so listen, I think our strategy against [Dell Technologies] has been totally opposite. They want to get bigger, scale. We want to get smaller, leaner and more focused. And I think, you know, we'll see. The jury is out, but the reality is what customers want is fast innovation and simple architectures they can deploy. Not a complex portfolio.

That’s why we have done a remarkable job simplifying platforms, focusing on fewer things, making it better, but then continuing to bring that innovation that keeps us ahead. The difference between us and them is very simple: We have an edge-to-cloud strategy. They don't have an edge-to-cloud strategy. Now they talk about that edge. You know, it's amazing to me when I was looking at a presentation they gave for the IPO they talk about the edge, but they don’t have an edge [strategy].

But with VMware, we have a good partnership. They realize they need us. For [VMware CEO Pat Gelsinger], being independent and agnostic is what he needs. We always want to give our customers the best choice. So in the end we are a multi-stack company and a multi-cloud company because in the end customers have a choice. The reality is customers are not going to put everything into one cloud on the public side.

As you grow, how do you stay nimble? Because you are going to get bigger.

For us, it’s to stay true to our strategy. And this is something that Meg and I worked together on over the last 15 months before she left. So this is not a coincidence that we have the strategy that we have. This was the smoothest transition in two decades. And so we think about innovation in three parts: organic, inorganic and through partnerships. Obviously, we want to invest organically. At the same time, we think about the inorganic plays that add to our portfolio to accelerate our strategy. We’ve done several acquisitions. You know … Cape Networks, Niara and Rasa Networks. And we will continue to do that when it makes sense.

On the core business we have SGI, Nimble, SimpliVity. On the services side, we have Red Pixie, CTP [Cloud Technology Partners] and Cloud Cruiser. Stay focused on that and make sure you integrate that in a platform. And then ultimately get efficient at what we do. And that’s why we put in place HP Next, which at the core is to rearchitect the company.

What impact will HPE Next have on time to quote for a product or service?

I've been with many of the partners, and to quote some of them, it cannot take more than five or seven minutes. You want instant gratification, instant pricing, and all of that. And it's not that we don't want to give it to you. It's just that the systems are not architected for that. In fact, I think we just retired a system which I think many of you know has been working at the company for 20 years. So I retired that system now. So I know we're still transitioning, but that's going to be good. But it's all about how we deliver against the needs that customers have today, which is speed. It's important. And obviously, from a productivity perspective, it's all cloud-based now and mobile-first approaches. So our reps and our partners need to have the ability to quote on a mobile phone. And so we're going to get that next year for the first time.

But at the same time we went through a massive systems process transformation. So what that means is that we've taken the number of platforms from 26 down to seven or nine depending on the product, we have taken 70,000 live SKUs now down by 75 percent, so you have to simplify the business to take advantage of this. And again, it's all about putting our people in a position to respond quickly, and therefore, improve productivity, improve coverage, improve the ability to make money faster. So that's what it's really all about. And to provide them a better experience because, ultimately, people want to work in a company where they can achieve their financial and career objectives. And it's hard to retain people when you have a system that doesn't work that way.

How has the innovation culture of HPE changed with you coming in as CEO?

This company has never stopped innovating. When I was running the compute business, the server business, I made three bets in 2015. One was HP Synergy, which is now out. The second was HP Apollo, which has been incredibly successful. By the way, Synergy now is a billion-dollar-run-rate business for us. And the third one was a platform for in-memory solutions. Today, 55 percent of the SAP Hana umbrella runs on HPE gear. So when you think about data analytics like Oracle SQL, even Oracle runs on HPE better than Exadata. I think for us, we never stopped innovating.

What we did, though, is connected all of that innovation together better, and we still have work to do. So a great example is that tomorrow I'm going to spend the whole day with my engineers going through that innovation top to bottom because we still have an opportunity to bring it together better in a common experience. And I'll continue to simplify and continue to make a few platforms that are drivers. And this is what I learned from Keerti and the team. So Aruba has that philosophy. It's a whole platform idea. So you can focus on Aruba Central, focus on ClearPass and Meridian, and now he's bringing together a cloud-enabled solution.

For us on the core side of the house, it's going to be HPE OneSphere and HPE OneView, and the rest is all enabling that platform to be the experience pane of glass we provide to our customers all the way to the connectivity to the public cloud wherever you're going.

But more and more, you're going to see a focus on software and a focus on AI. Our view of the future is that whatever you deploy on-prem has to be autonomous in many ways. And so we have this concept of AI Ops that we're going to deploy, and that's acquisitions like Nimble, which brought in InfoSight, and now InfoSight is the back-end platform for management of infrastructure so we can do predictive analytics and solve problems before they happen. And now 3Par and Nimble are soon going to be available on hyper-converged, and the rest of the infrastructure by the end of 2019. And if you lead with that, you have a better conversation with customers, because then what sits underneath is optimized to the workload, not the other way around. Because you have to start with apps that are not infrastructure apps. And that's the biggest shift collectively we have to make. And that's where I'm really focused right now with my team. And that's where we talk about data management, not just storing data. And that's what customers are looking for.

How should we be thinking about security, and how can partners drive revenue?

Security is a very complex topic. We think about it embedded in our solutions, and not as an afterthought. So it starts at the infrastructure level. And then as we move up the stack, how do we embed the security levels that are needed with encryption and everything else that goes with it? The more we go to memory-driven compute, everything in that transaction is going to be encrypted. You have an unbelievable opportunity here. When we sell our memory-driven compute platforms by 2020, every aspect of that is already taken care of, it's all encrypted at the core, from the app down to the infrastructure.

We still need to have partnerships because it's very important. In the case of Keerti, he opens the APIs to the platform to Palo Alto Networks to support because networking is a whole different game. But that's why Plexxi brings also a security aspect in their data fabric. So the way we're thinking about this is building that security within the solution embedded at the core, from the infrastructure to the software layers, all the way to exposing them to the app. And as we go along, we're going to teach our partners how to sell it. But that's one of the aspects of the road maps we are driving. Keerti has already done that in Aruba with ClearPass and Niara and building more AI, and he keeps adding, particularly as the platform becomes the platform to connect things. In the Internet of Things, you have to do it at a massive scale. But also figuring out that you need some partnerships along the way. But we want to minimize the number of things you have to do on top of that as we go along.

How big of an advantage is composable infrastructure for you?

We learn from the public cloud, right? So everything is disaggregated, and everything is presented to you as a simple experience, right? For us, we're trying to bring that public cloud experience and economics on-prem. It's the first infrastructure on the planet that compute, storage, network and fabric are totally disaggregated from the software layers. And through the software intelligence, we actually can compose from the app down to the infrastructure the exact amount of resources that an app needs to run, whether it is a virtualized workload or whether it's a container workload. And it all runs concurrently. And one of the problems we're trying to solve is utilization. So we can actually maximize the utilization of our platform to the extreme. Because when you go to the public cloud, what do you do? You go to the console, you log in, and you say, 'OK, I'm going to spin this VM, and I need these four things.' And you rent it. But you forget to turn it off. You turn it on, but you forget to turn it off, and therefore, the clock keeps ticking.

In our case, on-premise with Synergy, we do the same thing, but actually, you don't have to do it. The app through the layers can do that, and ultimately, compose and decompose. When the app doesn't need to be there because the resources are idle, don't keep those resources tied. Give it back to the pool so somebody else can use it and give it back to the pool. So we are composing and decomposing at the maximal layers. And because it scales to 400 nodes, then you're talking about big, large data centers at scale.

And so we can actually provide a block of IT with private cloud implementation, where you just provide APIs and people can still write on it. And a great example is from the developers. They don't care about provisioning infrastructure. In fact, they don't want to be slowed down by that. That's why compartmentalizing is key. They can get the things they need to develop their stuff—whether it's [open- source technologies like] Kafka or whatever it is—grab it, the thing composes, and then give it back. That's why we were so excited about this. Truly the first composable infrastructure. And with Plexxi now, we're going to actually compose all the way to the top of that.

So, therefore, the biggest problem that I have today is actually the networking, it's not the compute or storage. And why don't you treat networking the same way as compute and storage? And then you can speed up the deployment of infrastructure and services, even on-prem. So we are excited about that, and you can see more innovation. In fact, in six weeks, we will announce new things. So we are three years ahead, we're proud of it, and I think when the new technologies like memory-driven come online, we are going to have those available too. Actually, it's just a simple swipe.

How have your relationships with Intel and AMD been affected by the move to memory-driven computing?

All of these relationships are critical to us, and the fact that we're thinking about this memory-driven compute doesn't mean that the CPU doesn't matter. It's just that they play a different role in our architecture. We still need them to come to that data to process that data. We're talking a consortium with many of those partners in it. That's the interconnect fabric that connects the data architecture to the CPU. So that's super, super important. We want to offer you, our customers the best solution that's optimized for the workload. And I think about some of the virtualization aspects we see—in some cases, one of the CPUs is better optimized for those use cases.

We have a great relationship with Intel, with Cavium. With Cavium, actually we are doing now one supercomputer with ARM64-based architecture. Two years ago, nobody would have thought about that. But it was perfectly designed for that specific use case. And obviously, AMD has made great strides coming back in the data center with the new architecture, and the road map looks encouraging. But listen, we still have the edge aspect of this, and our edge line is all Intel-based. We actually can run an unmodified Azure stack at the edge today. So you don't have to drive all of that connectivity back. We can take that Azure stack and run it at the edge as is. So there are different things. In the end, it takes an ecosystem of partners to deliver against the needs that customers have. But the relationships are strong, and listen, Intel is going through their journey. Obviously, they've had some challenges as of late, but I'm confident they're going to resolve it. And this quarter has been a challenge for everybody. Look at the PC industry as well.

What would you like to get more of from the channel?

We have so many channel partners around the globe. This company was built with our channel partners. In fact, some of the channel partners have been with us for decades. They started with us, and we're super-proud of that. I think it's important that we continue to enable them to satisfy the customer needs at the pace we have seen before, but also work with us to make the right investment as we pivot. Because the pivot is happening. Whether we do it or somebody else does it, it's going to happen. I think for us, it's, ‘Come with us on that journey. We can give you the true end-to-end architecture that customers are looking for from the edge all the way to the cloud and the services and the enablement to go through.'

We didn't talk a lot about GreenLake, but I think GreenLake is a competitive advantage that allows you to position your services with our services together in front of the customers and give them a consumption-based model that they are looking for because they learned from the public cloud. But in the end, we need you to come along the way. And what we are trying to do with the new channel Partner Ready Program is to continuously provide that, makes the economics attractive, and also the simplification with certifications. Because we know we'll have to make it simple to make that pivot. We are incredibly grateful to our partners.

How important is innovation?

Innovation is priority No. 1, and culture comes with it. My three priorities as CEO are very simple: It's about culture, our innovation, and our execution. On the culture side, we now have 60,000 employees that have a totally different perspective than maybe 15 months ago. And I see it because I go everywhere. In fact, last week I was in Singapore, I was in Sweden, I was in Germany. You can see it, and the feedback we got is much stronger than ever before. We are really focused on the culture to the extreme. And to give you a perspective, we're going to move to a new headquarters, and I'm personally designing the interior because I believe that's how you impact the culture, the way that people work. The second part of this is always the innovation. And a great example is tomorrow. Spending a day in my jeans and my sneakers just focusing on that road map, because it's important that we stay ahead of the game. And the third one is execution. And that's why HP Next, or the Partner Ready Program, is all about improving the way we engage with you and making it easier to do business. Those are the three things that I'm really focused on. In the end, I'm really focused on customers and partners. That's where we are growing.

How did you become fluent in four languages?

It's very simple. My parents are both Sicilians, they were born and raised in Sicily and immigrated to Argentina. I was born in Argentina, so I did most of my studies in Argentina. And then I met this girl, and I moved to Holland, so I had to learn Dutch. And then when I joined HP in 1995, I was asked to support servers and routers and bridges in Italian and Spanish. But my wife speaks six languages, so I'm still behind. It's a conversation topic in which I lose all the time.

How stressful is your job?

Actually, it's not when you're convinced of something and you have a group of people working with you. I just spent an entire weekend with my staff and their spouses. We just brought them together for three days. We worked a little bit, but we also had a lot of fun. There is a sense of community and focus on the mission. And I feel very good about that. And it takes a team. Obviously, it doesn't take one man. The CEO job is the most lonely job you can find. But when you have a group of people that are committed, that they care about pleasing the customers and partners with a clear strategy, it's all about execution. So for me, it hasn't been incredibly stressful. Honestly. If you could figure out what's going on in the market—I'm talking about Wall Street—that would be a good thing. I'm convinced of what we're doing, and the important thing is to have the right work-life balance and have some fun along the way.

What do you do for fun?

I like to play soccer when I have time. It can turn out fine, but it can be very competitive. The thing is, when we go to play—I don't care if it's a friendly game—we play to win. And if we aren't going to win, we're going to pick a fight. This weekend, right, we had some fun, and then asked an individual to talk to us because we wanted to get a different perspective. And actually, I invited a five-time Olympian, 12 medals in swimming. And then she was nice enough to say that 'For those of you who want to come tomorrow morning at 8 o'clock, we'll do a class with you.’ And so she did a class. And so I was texting with her today, and I said, 'Listen, this morning, I had to roll out of bed.' So when I play soccer, my brain goes way faster than my legs. But you know the tricks because you grew up doing them. And sometimes, the weight is in your favor. Especially with kids. They think they're faster than you, but they have to go around you. And so we have a saying in Argentina, 'It's either the ball or the man.'

Talk about some of the innovative solutions HPE is bringing to sports including professional soccer and Formula One.

I think we have done some incredible things. But it's all in the context of innovating. And so, the clubs of Ajax and Tottenham are great examples. Tottenham is trying to open up a new stadium and has been delayed a little bit, not because of us. But it's going to be a digital experience. When you show up at the stadium, they're going to recognize you, they're going to give you a first-class experience, the whole thing that comes with us. And it's all powered by Aruba, the entire LAN and wireless Aruba. And if you go to the second level of the stadium, the cloud is actually sitting there, and it's all HPE. If you go to Formula One— talk about the edge. Think about it—it's a car moving 200 miles an hour, and that's a fast edge. Safety is No. 1, and No. 2 is performance. And so, we are doing a lot of work with them on data analytics, on the strategy, and on the safety aspect. It's fun stuff, it's a story you can tell, but the reality is it pushes you to innovate in ways we haven't thought of before.