 Live from Las Vegas, it's theCUBE, covering HPE Discover 2017, brought to you by Hewlett Packard Enterprise. Okay, welcome back everyone. We are here live in Las Vegas for HPE Discover 2017. This is Silicon Angles, the CUBE, it's our flagship program. We go out to the events and extract the signal from the noise. I'm John Furrier, the co-founder and co-CEO of Silicon Angles with my co-founder and co-CEO Dave Vellante with Wikibon and our next guest is Alon Andrioli, who's the Senior Vice President and General Manager of the DCIG, the Data Center Infrastructure Group at HPE. Great to see you, welcome back to theCUBE. Thank you, it's a pleasure to be here again. Great show, you guys have a lot of great innovations. Notable was the Analyst Press Conference that you were feeling all the questions around the buzz around Gen 10 and all the action you guys are putting inside the new servers from security to all the innovation that's happening. Pretty great opportunity. And the true private cloud numbers coming out of Wikibon are showing fastest growth is cloud on-prem. This points to significant opportunities. Your thoughts? Yeah, well, the need for compute is clearly growing and it will continue to grow forever. What we see is that the compute points are also expanding, so it can be on-prem, it can be off-prem, it can be in the edge. And on-prem, there is a bit of a revolution which is coming from the experience of the public cloud. And so private clouds are becoming very, very fancy. So you see on-prem compute basically turning into two families, very specialized for high performance computing, for mission critical, for AI and others. So things that are really very critical to the business. And then for all the other workloads, they need flexibility like a public cloud, but on-prem because they can keep control, they want to mimic the agility and they want to have the same economic level. So we are playing on both fronts. We are doing very well on the specialized front with HPC acquisition of SDI and so on. And we are making a breakthrough on the private cloud with Synology and soon with the new stack. So the whole notion of DevOps and cloud has opened up the doors and certainly you guys have been very clear with the simplicity message. Big data is a big part of the application process. Cloud providers, multiple clouds. So this right-mixed conversation is what Meg is putting out there is a nice message. And what you're saying is, hey, the on-prem is not going anywhere and we have the data to prove it, but you look at the big clients, they want the control. What is the conversation that you're having when they say, hey, I need more capabilities? Obviously high-performance computing, powering AI and machine learning was seeing obviously those things. But on the business model side, what are the customers asking from you for solutions? What are the key things that they want from HPE right now? What is that? In terms of economy control? Solutions, they're top priorities. When they sit down and say, well, you know, I need more compute. Okay, what does that mean? What specifically are you building for customers to help them with the digital transformation to simplify the business model of on-prem with cloud and to deal with the multi-cloud world? So they believe that the management of the mix between the different alternatives that they have right now is certainly a complexity and they rely on us to take this complexity away. So we are very bullish about the project, Newstack, because we think that this will allow data to be managed across the different horizons in the data center across multi-clouds and with more and more data being created and eventually computed at the edge. So these three horizons together make intelligent distributed computing, which will be more self-tuning, which will be extreme data analytics, and ultimately this will allow customers to manage data seamlessly across everything. We think that this is kind of strategically where customers want to be. Then the way they get there depends. Some customers have a view which is just modernization of what they have right now. Some of the customers want to be more dramatic and run everything they have as if it was a seamless cloud. And then they have to decide the mix between on-prem and on-prem. Most of the customers, I was looking at what is actually making the public cloud, more than 50% are born from the cloud. There are people who never had a data center and may never have one until they grow up, because then when they grow up, they need one. That's what we have learned. For control, economy, expands, that's a cloud cliff. So more than half of the public cloud customers never had a data center. About 15%, 15, 16% of the customers of the public cloud are consumers. And then you have a small surge, which are enterprises. So that's the first thing to realize, right? The move of the enterprise is still pretty small. I was discussing with the largest system integrator in Germany yesterday. And their view is a German perspective, because here in the U.S. we have a tendency to believe that everything is public cloud or will be. The German view is totally different, for instance. So I think we have gone through a cycle, which has been public cloud heavy in terms of Markham, where the market believed that public cloud was going to be everything. And we are now landing in a reality zone where this mix is an opportunity for the customers. There are some trivial workloads that can go on the public cloud, but we see that on-trend remains basically what people are doing. That last point's really important, because you said, well, let's maybe third as enterprises in the public cloud. If you look and feel the workloads that are going to the public cloud, it's not the core of enterprise IT workloads. So what I believe is that we are thinking it the wrong way when we're thinking public cloud and which workload goes there. The workloads are not going to the public cloud, is that a lot of the workloads that used to be run on-prem are now coming from the cloud. SAS. Right. That's different. That is very different. So customers are not deciding what is on-prem or off-prem. They are now looking at software packages that come from the cloud, like Salesforce or others. And this means that what they run in their data center are vital applications that don't come from the cloud. So it's more and more specialized. And then they have a variety of applications that don't come from the cloud that they will run on their public cloud. This is why I see these two typo topologies if you want of specialized, super compute and data centric, and then very fluid. And this is where synergy plays so well because synergy allows this fluidity of pools of resources and you can basically adjust to the various applications that you have. Oh, this is classic early adopter kind of behavior. You mentioned the SAS coming in and being influenced because they're easy to get into, right? You can get some subscription and get some value. But then I think the true private cloud is interesting to me because what it really shows to extend your point is that the business models are changing for the agility piece. That's the DevOps. So as you see IT consumption changing to cloud-like or true private cloud, essentially that is an OPEX business model. So the business transformation is now where the rubber hits the road for what digital is. So to me, we see that dynamic. So with that being said, what aspects of HP take advantage of? You mentioned synergy. What else do you guys have cooking up? What's out there that customers are using to turn the knob and go faster on the acceleration? But customers, I wouldn't like us to look at customers only as being enterprises because as more and more business is being generated from the cloud, people who do business from the cloud, whether they are enterprise service providers or software providers or business born from the cloud, these people also acquire technology and they have need for services and they acquire infrastructure. So this is a segment of the market where we're going to double down in the future. So we are looking, we call them like tier two, tier three, because the very large ones have a tendency to try to build their own things. But- Service providers, basically. A lot of other service providers and they are, they are a small sort of the market. Also demand technology and support from people like us. So we are going to expand our cloud line strategy. We are going to offer open systems and be very aggressive there, both for compute, storage and for networking. And so there are kind of two parallel market, if you want, more and more. There is a market of completely open systems. You call them white boxes. We call them for the cloud, cloud line, which is now a multi-billion dollar business for us. And then you have the people who want products that offer a lot of value that are differentiated like synergy, like ProLiant, like Blade Systems, like TripR, like Nimble and so on. And obviously we are doubling down on this as well with our acquisitions and on development like Gen10. So the narrative from Hewlett Packard Enterprise and some of your, all of your competitors is, you know, hybrid is the reality, fair enough. That's for sure. And we agree. But there is an aspect of zero sum game here and that the markets at the macro level are not, you know, growing like they used to. So market share becomes very, very important. You put up a slide on your keynote, 81 straight quarters of leadership. Now, we all know that you can play games with the numbers, but the most important metric we would argue is revenue share, right? If you're number one in revenue, that's the true mark of leadership. So you've had 81 straight quarters of leadership as we've just defined leadership. That's 20 years. In this quarter we had leadership and next quarter I think we will have leadership as well. But you know, we are not looking at market share for the sake of market share. We want to bring value to our customers and to our shareholders. So if there is moving forward a part of the market that does not yield value for either party, we may not want to measure our market share against that because we may not define this as being our own market. But so far, we are leading the overall market in compute. We are now a strong number two in storage with the acquisition of Nimble and we are happy to be there. But our strategy is not being number one for the sake of being number one. Right, but- Well, Jay's point, I'm very critical on this. I've been written about it and again, I may be overstepping my boundaries here, but I believe that if we're going to a new era of modern computing, the old metrics don't apply because everyone seems to be number one at something. I mean, I'll go to so many shows so I go to day where I'm number one in this. So the question is, if the old is shifting to a new model and it's horizontally scalable, vertically specialized kind of marketplace, which you guys are addressing with some of your tech, what are the metrics? So that we are asking ourselves the question, what should be the benchmark standard? So I have a strong point of view and I was discussing with it and it's last night we had dinner and I've had the same point of view for the last couple of years. The history of the market is to measure by product category. Rack, powers, old flash arrays, disk arrays, mixer arrays, and so on. I think that this is really my worldview. It doesn't matter. The decisions that customers are making are, what is my specialized computing, which includes computing, storage, networking? What is my specialized data center, basically? What is my private cloud? Then what is my consumption of IT coming from service providers? And therefore you have the service provider market which itself can be separated into different segments. That's the way to measure the business. So I want to be leader in specialized compute. I want to be leader in private cloud because this is what enterprise will be consuming. I want, and basically we already leaders there, but I want to continue to be leader in providing gear to service providers who have decided to rely on partners to build their data centers and not build them themselves. We want, this makes sense because then you look at the market differently. You're not looking at lack or tariff, density to optimize, whatever you're saying, okay. What is a service provider going to need in the future? What is going to be specialized computing in the future? What is going to be a private cloud in the future? Once you have covered that, what is going to be compute at the edge in the future? And what do you need to orchestrate all the data? These are the clusters of the market that matters. They are the one we are pursuing and they are the ones of which we will be the leaders moving forward. 100% agree with you. I would also add, by the way, I agree with you, 100%. And I would even amplify by saying you could be building something new like a server, chip, silicon, security. That has no category. So how does that relate to the thing? Well, seniority has no category. Right. You know. It's horizontally scalable. So again, you could be number one, two or three by the old categories, but be holistically number one in the market. So I think it's more, you know, it's more categories of business outcomes. Like specialized high performance, you know, flexibility, agility of a private cloud and so on. I think that's, you know, so if you make a parallel with the car industry, you could say, is the market like diesel engine or gas engine or electric engine, or is it like sport cars, SUVs or whatever? I want us to look at SUVs and sport cars. How do we do the best SUV? How do we do the best sports car versus, you know, the components, this technical view of it's a rack or it's a tower? How do you add the most value for customers that is profitable for shareholders? At the end of the day, when we have our argument in our office about this on the research side, we say, look, at the end of the day, let's identify some of these new categories, measurement points, but customers and revenue can't lie. If you have customers, here it is, number of customers. And so the problem then is to measure it. Once you have defined what is the right metrics, can you measure it, right? And so unfortunately, the analyst today cannot measure the market, the where it has evolved. So we are still looking at rack and towers and so on and I think this is the wrong. Okay, so talk about the hot thing that we like is the root of trust product, the silicon thing that's called the root of trust, you know, with the firmware thing. This seems to be getting a lot of buzz to show. It's innovation. We had some independent testers on with your guys on the Gen10 sites and this is pretty impressive. Thoughts on, is this the kind of direction you continue to go with? What's your thoughts on this security? Yeah, we think security is super important and you open the newspaper or the TV today and you see what's happening. It's quite amazing, including today, what's happening today here in the US. So it's incidental that we're coming just today with our new generation of compute but it's taken two years of interviews with customers to really understand what's most important to them and the risk of cyber threat has turned enormous and I think that you have been interviewing experts from the FBI and so on during this session who came here and helped us to build this solution and I think we're coming at the right time with the right solution that will take a few years to our competitors to try to match that and they will go in this direction because that's the only way technically you can do it. It's at the silicon level. So you basically have unique encoding on your server in silicon and the firmware always compare itself throughout the whole life cycle of the server even before the server is finally built to this root of trust. I think we've done this extremely well. I'm very, very proud of our engineer. And it's been validated against the NIST NIST security standards so congratulations on that. And these are the most stringent standards in the industry, right? It's pretty impressive. I mean, this has been a trend that we've been seeing the silicon angle, no pun intended. It's interesting, it's always on securities come up in the past, people want that. And with IoT the support, the attack vectors can be sealed up pretty well. And so our edge line products will have ILO 5 and so they will also have access to this technology. Great innovation. Elan, thanks for coming on theCUBE. Really appreciate you sharing the insight. I'll give you a final word here. Share with the audience something you think they should know about HPE right now that they may not know about. I know the messaging is pretty simple. You got the nice messaging, but going beyond the messaging, what would you like to share with the audience about your group and HPE's innovation coming out of Discover 2017? You feel the buzz here. You can see, I think we have never been in such a focus and clear position. We exactly know the businesses we are pursuing, the hybrid IT, make it simpler, and the edge and the service to make it happen. We are just crystal clear. But when you put the tree together, you get to this dimension of intelligent distributed computing. And this is a market that we will lead in the future. Also, we are such a strong and stable company. We will have over 12 billion dollars of cash net in our balance sheet by the end of next month. And this put us in a position to continue to double down on these bets we have made for the future of the market. So we are very, very confident that we are in a great spot. And frankly, it's great now because it feels like we are starting to be at destination. The last 18 months, we separated from some of our legacy friends. And now not only are we on our own, but we have a super clear strategy moving forward. We are proving that we're implementing it with the six acquisitions that we have made over the last few months, and more in the pipeline, continuing to deliver the capability to integrate this acquisition and the capability to continue to motivate our customers to be with us. And the spotlight is on you guys. We'll be tracking it. Thanks for coming on theCUBE. Really appreciate it. Senior Vice President, General Manager of the Data Center Infrastructure Groups, sharing his opinion here on what's happening and where's it going in the future for HPE. We're back with more live coverage for theCUBE here in Las Vegas after the short break. I'm John Furrier with Dave Vellante. We'll be right back. Stay with us.