 Okay, we're back in Las Vegas live SiliconANGLE.tv, SiliconANGLE.com, Wikibon.org is the cube flagship product that we go out and talk to. All the experts, thought leaders, executives, CEOs, startups, you name it, here we're covering the Yulipacker Gen 8 launch, extracting the signal from the noise. I'm the founder of SiliconANGLE, John Furrier. I'm here with my co-host. I'm Dave Vellante of Wikibon.org, and John, we're really getting into it now. We're starting to hit our stride. We're here with Mark Potter, who is the Senior Vice President and General Manager of HP's ISS Division Industry Standard Servers and Software Mark, welcome. Thank you, really glad to be here with you guys. Great, first time in the cube, so we'll be easy on you. Well, relatively easy. We just talked with Dave Donatelli, who gave us a great overview of some of the market trends and some of the segments. Obviously, you're running the whole project here, was up on the keynote of the press event. Gen 8 really is part of a bunch of innovation around project moonshot, which we covered on the cube, the low energy thing, was Odyssey for Mission Critical and that world here. This is kind of the mainstream market. This is the bread and butter of HP. Right, right. And HP's not, everyone knows HP for servers, right? Yeah. Everyone's got HP serverization networking. So you're known, but racks and racks of servers are out there from the big companies to the SMBs. What's the core problem with that Gen 8, this new innovation that comes out with? What are you seeing as the core problem? So fill in the blank. The core problem is blank. The core problem really is manual operations and the lack of automation in the data center. I mean, when you think about, when we went out, we literally talked to and interviewed hundreds of thousands of customers over the two years. We looked at trends. We looked at outage events around the world. We talked to partners. And the number one thing they said is, is HP, please help us on our manual operations, automate, build it in, make it simple. It's too difficult. We're spending too much time really managing day to day operations. Too hard to provision and too hard to even do basic updates. So two years ago, $300 million later, great vision, took a risk, as Dave said, and kind of it worked out, but the market kind of turned around too. The IT economy was kind of stalled for a bit. Everyone's kind of a bare bone operations cutting down to the bone. Now it swings back on the other way. Top line revenues of big driver applications, tsunami is seeing it on the consumer side. So you see in business drivers, top line revenue, top line revenue. And on top of that, cloud computing, and obviously Moore's Law is always in effect, big data. So talk about how that all came together. So in your two-year period, doing all the market research, get everything right, dotting your eyes, crossing your T's, and all of a sudden the market lift comes in. How does that affect you guys? Well, I mean, it's important because anything of this kind of groundbreaking, architectural significance takes time. And we're in an industry where you've got to make bets. And the trends we saw were really two-fold. One, as I shared, is that manual operations in the data center and stranded capacity in the data center were really two big issues. Big issues where admins weren't spending time really getting to the innovation and their businesses wanted them to. And so what was happening is the businesses were moving past IT in many ways. And so IT really needed some help. Second thing is with the whole energy situation we have, when we looked and saw all the stranded capacity in the data center, we knew there had to be a better way. And so we really addressed both of them. The fact of the matter is the third big thing that we saw with respect to all this is just the incredible growth of the data being stored with unstructured data that you both talked about, big data in just the scale of the cloud. I mean, more and more customers that we were dealing with every day weren't talking about 10 servers, they were talking about 1,000, they were talking about 10,000. And so we said, why can't we bring that simplicity and scale across the board to every single customer? It's just the timing is so good on the demand side. I mean, just when you thought all the prognosticators talked about market share, here's the demand, all of a sudden with the data tsunami and these requirements at the edge of the network, which you guys know a lot about, right? So congratulations. And so, David, you want to ask a question? So Mark, you've seen a number of transitions in the industry. So my question is, previous transitions have been driven by whether it's microprocessor performance or maybe it's storage costs, lowering, what's driving this latest trends? What is that transition? How would you describe it? What's driving it? Well, I mean, those things are still really important, but I think the transition this time is really different. It's really about how do we get to scale and simplicity and deal with the scale in a fundamentally different way? How do we deal with the operations of the data center and really dramatically simplify and fundamentally deliver shared services to the internal customers? And in order for IT staffs to really do that, they've got to be able to have systems that do much more of what I'll call basic operations work. They've got to have systems that have the automation built in and that fundamentally, just like we've done, take care of themselves so they can get on to delivering kind of shared services and move much faster because the one trend that I see more than anything is the pace of change has picked up dramatically. The economy, as you said, is coming out of 2008, 2009 slowdown and it's not just about the economy lifting, it's about the amount of change that seems faster than ever. Data explosion, the different services are available and IT's needing to move faster. Donatelli was talking about John when he was on and marked the big guys, like the big web guys, Facebook and Google and of course your people, the traditional data center. Do you see those as a leading indicator, those being the Googles and the Facebooks? I mean, we saw that with Project Moonshot, the hyperscale was sort of really geared at that audience. And I said to Dave, well, and even Voyager in a way is the trends are really driven by the Googles of the world. And really what I meant was those guys have been driving automated operations for a while. Do you observe that and say, hey, this is an indicator of things to come? Or is it more you're on a different parallel path? Well, I think it's an indicator more than anything and I think it's back to kind of watching those indicators and saying the business model is different because where let's say Google or Facebook maybe has kind of one application across the data center and I know I'm dramatically simplifying because their IT is much more complicated than one application. The enterprise has thousands of applications and the enterprise needs the simplicity of the model that they enjoy, that they need to be able to do it and deliver all those applications. And so with the intelligence we've built in here all the way up to our cloud system product, what we've really allowed the enterprise to do is to have the simplicity of a shared service just like those customers and those indicators but we're doing it in their space allowing them to really take all the applications they have to deliver for IT and do it in the scale and the simple way that the cloud error demands. That takes a lot of hard work. That takes a lot of hard work because of the diversity and complexity. Yeah, we have some guys in Santa Clara covering the cloud, small little public cloud conference, cloud connect and obviously that's the cloud service providers and that's the emerging kind of the public cloud guys. We've talked with SAP, we've talked with HP and all the big guys in that business. They're pretty, it's a battleship as I say, right? So, but one of the hot things in the emerging side of this market is DevOps, developer operations where it's a developer centric model. So you mentioned applications. How do you guys fit into that mindset? It sounds like you're treating it like the lower end of the stack, providing core infrastructure, kind of Lego blocks as Dave Donatelli says, but what's your vision for the DevOps world? Because we've had many conversations, contentious ones with guests on theCUBE where, well, DevOps is BS. It's actually ops dev. Developers don't run infrastructure. And if they did, there'd be no tolerance for failure because developers have bugs in their software. So, you know, there's a whole dynamic going on but the developer frameworks are expanding, you know, VMware, what spring source, we're seeing Hadoop. So, more and more developer centric views are out there. What's your vision on that? Well, I think it's a critically important area. And you know, what HP's all about is we think, you know, ultimately our cloud and the cloud customers needs are all about kind of hybrid delivery. And that cloud has to be linked across what I'll call private cloud on premise to maybe even managed private cloud where somebody's doing it on your behalf all the way to public cloud. And when you're talking about DevOps, I think that's a critical, that's what I'll call one of the most critical use cases across this hybrid cloud. Because if you think about it, developers may want to actually develop part of the application and test it in the public, but they want a seamless way to migrate that application and that service right into the enterprise where the data is protected, where it's secure, et cetera. And I think that's going to be an incredibly important use case. And it's got to not only map to the private cloud deployment, but also across this hybrid delivery where maybe part of the development is actually done in the public cloud. But it's a, and I also say it's a continuous agile process. It will be done continuously and you will have to deploy some of that application in the enterprise, in a private cloud and some of it may actually show up in the public cloud. It seems to me that the market's going and I guess on a follow, we first of all, we love DevOps angle. In fact, we just launched DevOps angle vertical that we're expanding coverage into because we see it as very relevant. I think it's one of the biggest use cases that's out there that needs solving. And it's emerging. So, as you say, things kind of incubate and then get small little communities developing. We just covered the Node Summit and Node.js as how one of the hottest frameworks around IO and these web apps that are being developed. What inning would you put the DevOps in? If you had to put it into a ballpark terminology. We're early innings. First inning? We're early innings. Third inning, early, very early. Yeah, third inning. Patty practice. Yeah. We're in the on-bank training. We haven't made the cuts yet. We're in early, but the whole, what I'll call DevOps, you know, HP has an incredible amount of assets here on application life cycle management, you know, obviously with our, you know, quality center, et cetera. And so, we have more experience here than probably anyone in the industry. And so, while it's still early, I still think this is an incredibly important use case. Okay, I'd love to follow up with you later on that, but that's great stuff. I wanted if we have time just to go into a little bit to your keynote. Yeah, huh? Just to share with some of the fun facts. He said $24 million spent on manual ops over three years. 10,000 square foot data center, right. $29 million energy cost over three years, unplanned downtime cost, you know, in some cases $10 million an hour. You're talking about Voyager driving three X the admin productivity and then giving 30 days back to admins. Yeah. It was interesting, the analyst from Forrester said, and I think he was joking. We gotta watch out because, you know, we're gonna be losing jobs. Yeah, Rich, I think was joking. He was joking, but do you ever get that kind of? Yeah, you know, sometimes, but in any transformation and any kind of big step that the industry takes, and these servers are a big step forward. You know, there's a people side to that change. And, you know, I think as we talked about it, when we've shared the generation aid and all the innovations we've put into this proactive insight architecture and what the project has delivered, admins are like, you know, it's, they're celebrating. Yeah, yeah. Because, you know, they can do more. They can do more exactly. They're creative. They're not laborers. And it's like that human capital that is, you know, spending time on doing basic stuff like firmware updates. Yeah, mundane. Where they get fired, by the way, if they screw up. You know, it's like what I showed, it's like, how long have we been delivering white papers to the industry for optimized configs? Why can't we deliver a mobile app and take that white paper and have something as simple as a QR code and automate that whole thing with that intelligence? Mobile app, I want to apply this optimized configuration to that server plus a thousand others. So this is great news for you server admins out there, you know, update the skill sets and start making applications run faster, delivering more business value. That's where it's at. And I agree. I think that is exciting for that class. I just was telling John, I wrote a piece on the Wikibon Wiki, how storage admins can remain, or you can remain relevant. And I think there's a server admin analog here. And this is what you're looking at. Yeah, we've seen the same thing in the Oracle world where the DBA is moving from that title to something much broader to a scientist and or more specialized, not some database administrator. So seeing a lot of roles change in IT. So that's exciting. That's good news actually. It's great. Sometimes it's hard. Change is good. Change is good. Okay, we're here with Mark Potter from HP, Senior Vice President of the Gen 8 Project here. Thanks for coming on theCUBE. We really appreciate it. Thanks for your time. Thank you guys for the time. My pleasure. My pleasure. Thanks for having me on theCUBE. Okay, first time on theCUBE. We now have another alumni to put on the list with Jim Gontier and a slew of other HPE execs. That was good. Another first time CUBE person. So he basically validated the David Fleurer's assessment of the core problem, which was operations. Okay.