 Live from Las Vegas, Nevada, it's theCUBE at HP Discover 2014, brought to you by HP. Hey, welcome back. We're here live in Las Vegas for HP Discover. I'm John Furrier. This is theCUBE. My co-host, Dave Olavde with boogiebond.org. Live in Las Vegas, HP Discover of 2014. Our next guest is CU alum, Naisar Goli, the Senior Vice President, Chief Operating Officer of the HP Cloud, Senior Vice President, Chief Operating Officer. Welcome back. Great to be here as always. So the team gets bigger, the cloud gets bigger. Just take a tweet of Bobby Patrick holding his hands out here. He was like, I caught the fish, it was this big. It's been great to follow your career. You built this from scratch. Operationally, it's got a great foundation. Now you're starting to put the walls up, hang the iron, so to speak, and start to shape this thing out. So here's the update on the operations of HP in the variety of different cloud. What's it starting to look like? The foundation set, I said like a hanging iron to use that metaphor. What's it shaping up to be? Well, like I said, we talked about this two years ago and we started building it. We were only a few people, but we had a vision. And we now have hundreds of people working on it. It was the launch of Hylian, obviously, we now have our distro out. We have cloud system, wizard technology as well. We have our public cloud. We're pushing it into Manitz cloud as well. And obviously it was the Hylian network that we just announced. So it's all coming together. I think that the beautiful part of it is this is really what we had in vision a few years ago, but you can only do this if you have commitment and you have funding. And luckily we were lucky enough to get funding from Bill and Meg and Martin to actually make it happen. And I think that's the amazing piece that HP is really investing in making this go. You guys had a great presence out in OpenStack Summit. It was impressive to see from a, not just the normal OpenStack press, just from a talent standpoint, the team's getting bigger. So before we get into some of the whole operational pieces about the team and the different versions, talk about what the different pieces of what's working today. You have a lot of different things announced. You have stuff that's in GA software. The big announcement today about HP's cloud, Hylian cloud for the service providers and everything. Break it down, break out the offerings. This is a lot of different use cases. Some shipping, some's coming in Q4. Break that down for us. Sure. So we have, first of all, we have and the Hylian really is a portfolio of products. And there's different pieces of that portfolio. First of all, we have cloud system. So there's our market leading system for building private clouds. And that is an unbelievable product line. It's like on a rocket ship. And that also has our OpenStack in it as well. Then we have the distro we announced, which is the Hylian Community Edition. And that's really for people to play with. You can download it for free. It's like a sandbox to get a feel for what we have. And then that's going to go GA later this year around August. And then we're going to have what's called our Hylian Commercial Edition or HP Hylian OpenStack. And that's really going to be the enterprise to raid high scale, heavily integrated for people who really want to build large scale clouds. We also announced this week, actually, our developer platform. And the commercial, is that available now? Is that going to come? No, that will be the commercial edition is going to be available in general availability is probably around August sometime. And that's actually not coming out. I think it's not actually called commercial edition, but it's actually the Hylian OpenStack. The current one is a preview, Community Edition. This one is sort of just the plain version. But that is the version that for customers to build clouds with. So wait, so Hylian and Hylian OpenStack. Hylian is the umbrella brand? Hylian is the umbrella portfolio. So Hylian is the cloud portfolio from HP. And like I said, that has cloud system in it. It's got some of our management suite in it. It's got our distros in it, which is like the Community Edition and the OpenStack. And it also now have a dev platform, which is based on Cloud Foundry, which we launched here and discover as well. Okay, so let's just get this. Cloud systems, distro and Hylian Community Edition, commercial edition, Hylian OpenStack for large scale, GA this summer. What's next? We have the developer platform, which we launched the Community Edition of that here. And the commercial edition will be later this year as well, which is our development platform for Cloud Foundry optimized for OpenStack. And then Hylian OpenStack is your OpenStack distribution, which I got to ask you about Red Hat, which is not certified on Red Hat. I don't know if you guys talked about this at OpenStack, but what's the story? You guys obviously would welcome Red Hat certifying, right? By the way, this is before that, I just want to say one more thing. We also have our Hylian Public Cloud, which is the public cloud we've had running OpenStack since 2010 or so. Which completes the portfolio. Right, sorry. Back to the Red Hat question. Look, our view is we shouldn't have this artificial connection between Linux and OpenStack. And that's one of the things that we've noticed as we've talked to customers, where some of the Linux vendors, you've named a few, there's others have tried to create this artificial thing where if you're using OpenStack, you have to use this version of Linux and therefore potentially you have to end up paying off kinds of licensing fees. We don't think that's really what people had in mind when they came out was OpenStack, right? OpenStack was not supposed to be open. And so that's one of the reasons where, when we put out our Hylian OpenStack, it comes in was a host Linux built into it, based on Debian, but we've optimized. And oh, by the way, we'd be happy to get that into OpenStack for everybody. It's not specific to HP. And so that's our view. We don't think that whatever is happening in the Linux world should be telling people, informing people what to do in OpenStack. That's the first piece. The other question, of course, is okay, so what do you use as your guest operating system? And our belief is you should be free to use whatever guests you want. There are certain vendors who believe that their guests only work with their host. I think that's somewhat monopolistic and I think customers have figured that out. So if I use a certain host as OpenStack, why should I be restricted with guests I use? That's not the purpose. And so I think the good news is there's a lot of the discussion about that in the ecosystem and these things usually solve themselves. So I don't want to, I'm not trying to serve the pot and I don't want to really rat hole this, but just for my own edification, you can run Red Hat's distribution on your OpenStack platform. They just won't certify it. That's right. Right? They just won't certify it. Now you want them to certify it, you'd like it, you'd help them certify it. Yeah, they've certified it on VMware and they've certified it on Windows. For whatever reason, they don't want to certify it on someone else's OpenStack. I think the question remains with them, why they would discriminate against OpenStack specifically. I thought they were big supporters of OpenStack. But you know, everybody's got to make their commercial decision. I would think that when you reach critical mass that they would certify it on your platform. Why wouldn't they? It's good business. I mean, unless it's, as you say. I think, again, the way these things usually work out, especially in open source, is let the customers decide with their feet. And ultimately they will. And I think what we've told customers is, hey, look, you have an ELA with Red Hat, you run what you want to run and let them tell you that you can't run it. And if the customer's big enough, that usually works out pretty well. It reminds me of Oracle on VMware. Same kind of fun. Everybody just said damn the torpedoes and it all worked out fine. I don't think there's any problem in this area. Talk about OpenStack for a second. Let's talk about HP. So you guys had a big presence there. What's going on there? A lot of buzz around you guys now. Now that you have more feet on the street, Bobby Patrick was just on. Now you got a marketing guy and you guys are going to turn up a lot of action. What's the status? We had Eileen, we had Monty on OpenStack. How many people do you have? What's the influence? What are you guys doing to lead the other six? Last year you said something big is going to happen this year in OpenStack. Tell us what happened. Come on, man, what's going on? Everybody kept on saying, where's your distro? So we have a distro. Now the reason we launched a distro when we did is because we felt that it was OpenStack was ready. Back to what you said, as you know, having followed OpenSource, the key to OpenSource is the community. And the part that I think has been most important is regardless of all these announcements and products we ship which are amazing, the part that I'm most proud of is that we've actually dramatically increased our presence in the community. We have seven PTLs, right? Seven team leads that are voted based on merits in OpenStack, that's a huge deal. And the reason that's useful for us is also because it helps us drive enterprise grade into OpenStack. And so really, I think the key thing, and also if you look in Juneau today, right, we are number one code contributor in Juneau right now. Now that could change. We'll be number one or two, that's not our focus, but so when we talk about being serious about OpenStack, you know, we're putting our money where I'm at. So that was really substantial. I mean, number one, number two is irrelevant. Is it meaningful? Is your contribution meaningful? Well, we have product leads. Some of the product leads. How many product leads do you guys have? Because how many, like 15, 16 seats? Or what's the breakdown of the project leads who lead the projects? Because that's the leading indicator. Well, like I said, we have seven, all right. I don't remember each one of them, but you know, we got triple low. Out of how many? I think there's 14. So we have about half, close to half. As project leads. As PTLs. Okay, so I got to answer the next question. As I smile and kind of think about how I want to say this. What's broken about OpenStack that you guys want to fix? So you got leadership there at a technical level. What is the urgent areas? You got the, there's some rooms in OpenStack that are on fire that need to be put out or rearranged, however you want to say it. Obviously there's a lot of demand for OpenStack. No doubt about it. What leadership are you guys bringing and what do you see me doing? I think, look, OpenStack is evolving. And I think one of the things that has been very informative to us is the fact that we've running a public cloud. So we've run into a lot of these things way up front. It's one of the, you know, when we ran on a public cloud ever since Grizzly. And, you know, look, I don't think I've already been, you know, been quoted in this, you know, networking needs work. It's not that it's broken per se, but it's not where it needs to be. And the good news is everybody recognizes it and we're doing some work there. You know, install was triple low is getting better, but there's work to be done. Monitoring work to be done. And so I don't think that something's broken per se. I just think that OpenStack is evolving and we shouldn't get ahead of our skis. But, you know, today you can deploy it, but we would like it to be more seamless, less involvement, clouds all about automation. You know, OpenStack still requires more involvement than we would like. One of the reasons, one of the things that we've done as part of our launches this year is actually we're building out a professional services team. They already have a very large team there and we ended up 100 professional services because customers have asked us, along with their distro, they also want us to help them actually deploy this. So we're building that out as well. But, you know, I don't think things are broken per se, but, you know, there's definitely opportunity to continue improving the product. And that services is part of your team, sorry, John. Yes, yes, they're specifically put under the cloud business that actually worked for me, but that's not that important. They're part of the cloud organization because in the spirit of DevOps, everything has to be tied together. They have to be next to the developers. They can be off somewhere in the woods. So what are you most proud of this in the past 12 months? Obviously last year, we had a great conversation. You were very bold. You were like, hey, you know, we're going to do some big things. I think we might have been Barcelona, but I can't remember, but it seems like seven, sounds like dog years. But now, what are you most proud of this year? If you look back, I think, you know, we've managed to execute through what our plan was. You know, our plan was to join our public cloud, get a distraught. Our plan was to significantly hire talent. You know, we've become the destination for talent. People who want to work on open source and have a free hand, an open stack, we are a destination. You know, we're getting tremendous hires, tremendous talent. We brought in talent in like Bill Hill. You know, great guys like Bobby, we got in carry, so we up the talent level. You know, and so I think we've become a destination for people who really want to push cloud forward and it's been able to grow the team. And so I think that's probably what I'm most proud of, that we really are driving from that perspective. So I wonder if I could ask you, I've been asking this question and I really don't feel like I've gotten a straight answer and I almost always get straight answers from you. So, there's experience between on-premise and the public cloud. Yes. How consistent is that today and what's the vision as to how consistent it will be and when? Well, I mean, what do you, in terms of seamlessness or what do you mean? Well, so you address seamism. I'm specifically thinking about things like security policy and auditability and the consistency of that compliance between my on-premise and my hybrid. That's a complaint that I hear all the time from people. It's just too hard. I negotiate with a vendor for six months. I'm comfortable with that. And then I turn to the public cloud and it's a completely new process and it's just, it's frustrating. Well, that's part of the reason that we're driving this hybrid architecture, right? It's not 100% there yet, but when we talk about having the Helian OpenStack, the idea would be that you would have the same on-prem and the same off-prem what we also discussed in the Helian network is the greatest federation that all uses the same architecture and has the same baselines. Because the key is to have the same baselines. So, I'm worried that outside that network I won't have interoperability across OpenStacks. Should I not be worried? Well, look, you should always be, I wouldn't say worried, but you should always be open. You should always look at the details. The devil's in the details. Everything is perfect until you get into the details. But I think as OpenStack matures and in this network you will be provided that. Depending how much closer you're to the network, the more it will be seamless. If you get further afield, it may not be as seamless, but it's still gonna be a lot more seamless being OpenStack than going to a completely different architecture. But the beauty of this network is it's a very inclusive network ecosystem that's going to have major service providers across all the globe. And so the chance of being out of that network is going to be very minimal. You'll be able to get all your services from that network if you so want to wish. Right. So it's clear obviously Amazon and Google have put the fear of God into many folks. What's your take on them relative to the business opportunity? Obviously right now it's not truly overlapping or conflicting other than maybe a potential collision course together in the enterprise and those guys. Look, I give these guys a lot of credit for driving this architecture. They did it from within. They built the stuff, the DevOps, the cloud, a lot of the things that are being driven. They started. But between that and mission critical enterprise grade, there's a long way. We're at the beginning of cloud and so they've started a great thing. But I think that while on the technology front they have some really great technologies when you get into enterprise grade and mission critical and the kind of things that enterprises expect. There's a question of, do they actually have that DNA? And I think that's where we have a big advantage. Can we talk about pricing a little bit? As you guys announced, a much simplified pricing model. I mean, Amazon pricing is very complicated and it changes every 30 seconds, which is I guess a good thing for consumers maybe, but you're pricing $1,400 per server, right? Describe what that means. Well it basically means if you want to run a private cloud or you want a public but a private cloud and again it depends on the size you want. If you want larger and you want to make bigger commitments it has to be cheaper, right? For $1,400 a year you get our full supported stack, tool and demification, everything, including the host Linux. All the services. All the services. I get object store, I get Cinder, I get everything. You get all the services in OpenStack. So, again, enterprises, OpenStack is not about price, it's got to be a competitive price, but it's about simplicity and so that's not complicated, right? Some of the other vendors will give you very complicated pricing when you actually add all the support and everything then it all of a sudden becomes kind of problematic. And we're also looking at other ways to price things. So we're going to be very creative in terms of pricing for use and so forth, even on-prem because we want to make it easy for people really to have the cloud experience regardless of what deployment model they're doing whether it's on-prem or off-prem. Well, the customers can turn around and monetize that, particularly cloud service providers or internal customers can charge for that. That's a nice model, that's a simple model for them to consume and then they can build out their services catalog however they want to do it. Simplicity is important. David and I always talk about sports metaphors, whether you're talking about Formula One racing or cars. From back to the endings. We look at you as a mechanic in the shop. You had some parts, you put some cloud together, now you have more parts coming in, the shop's getting bigger. So I want to ask you two questions. One, what have you learned in playing with engines, getting, rebuilding them, taking them apart, re-pitching them together? And two, what's your plan for the year? So first, with what you've done, which has been pretty much from the ground up, getting your hands dirty with the public cloud, to now, what have you learned? What scar tissue, how can you share with the folks? It's kind of a best practice with the look for, with the potential speed bumps, our intention to deal with public cloud and hybrid cloud. And then what your plan is going forward this year? Well, we've learned a lot of things. We've learned that now working doesn't work. Although we fix that. I think it's going to be important to really understand what are the key, when we started off OpenStack, us in the community, there were many different projects and lots of different things. And there's a lot of benefit to having all these interrelated projects could move really fast. But at the end of the day, the customers are consuming a system. And so we need to look at this more of a system and less of its projects, because from a customer's perspective, it's a system. And part of the challenge is you ask me, what's the challenge in OpenStack? It needs to look like a system. The monitoring, the networking, all of that needs to look like one system. So I think you'll be seeing us do a lot of work in that area, because that's the way the customer wants to consume that. And so while the way that OpenStack is distributed everywhere, has enabled us to get great speed of innovation, at the end of the day, it all has to come together as a system. And I think we, the community, have some work to do there. So it's not so much a flaw, it's more of a continuum of evolution, right? There's an evolutionary process within some holes right now. Well, you run really fast, and then you've got all these different things, and you say, you know, it'd be really nice if they all sort of worked together better, and so you say, okay, now we're at that stage, the maturity stage, you've gone from the adolescence to teenager, and we need to go to college, and you know, you go to college, you've got to be a bit more polished. So follow up on that. Where are the white spaces that are opportunities for other entrepreneurs and companies? And two, where do you see specifically the paradigm looking like something else? Meaning, you know, is it systems management? What does this look like from a legacy statement? Where have we seen this movie before? Systems management, is it networking management? I think when you went to client server for mainframes, right, it's something like that in a sense where the whole way you look at things is different. I mean, the DevOps concept, right, is very different for people to understand when they start, you know, when you go to someone and you say, listen, your enterprise app is going to upgrade every few weeks. How does that make you feel? Uncomfortable. Unless you're a DevOps chef. You know, when you talk to people about open stock and they say, look, I want six years support, I'm not like really, in six years, you know, whatever you're using is going to be up to lessons. Dog years, that's not a good idea. And so I think that is really, you know, changing, you know, so the concept, it's the same when you went to PCs, you know, and you know, mainframes would upgrade once a decade, and all of a sudden every six months you get a new PC, you know, new Intel chip, you know, because of Moore's law. It changes the way people look at things. The same thing is here where your cycle of innovation, you know, keeps on increasing. The speed of change increases. And you know, if you look at the thing what I find interesting, you know, if we look at evolution, right, what's sort of determined who has been successful and has really been who can handle change the best. Not who's smartest, not who's biggest, who can handle change the best. And you know, businesses like that and cloud is now like that too. And so it's accelerating. So if before handling change meant you could do, you know, three or four upgrades a year, you thought you were a hero. Now these guys are coming to say, yeah, I can do it every week. So what happens to those 19 year old enterprise apps? They just get away. Well, I think that's a good point. Look, it's like, you know, we still have mainframes, right? And so apps that are very dynamic, but things where there's a lot of change, a lot of shift, are going to have to get rewritten. So it's a balance, right? And apps that are sort of static and they're just like, you know, some records thing, whatever, they'll just stay and there'll be an API to call them. And they'll just, you know, you'll put them in a box and you leave them in that box because, you know, you want to be, you want to, you only want to rewrite stuff or write new stuff where you think you can get a benefit. Some things, again, there we still have mainframes for certain things, because there's no value in rewriting them. But you know, the thing that, but they'll be in a box. They're not going to be part of your new experience because your new experience, people's expectations are just very different. They're expecting, if I want a new feature, okay, can I get it next week? Right? It's not like, okay, yeah, I'll put it in in three years from now so the future will show up, you know. So how do you whip those old guys and train the new, how do you whip the new, old guys in the shape and train the new blood coming in to be DevOps? Because what you're really talking about is, that's the Mark Zuckerberg, that's the kind of culture where move fast, break stuff. But in the enterprise, you can't move fast, you have to move fast, you have to move fast and keep it supported. Which is now Facebook's wrong motto now. Well, I try to avoid whipping people in general, but we're tolerant, we're inclusive. I should talk to some of my engineering VPs about that and we'll have something. Bullying is not cool anymore. But I think we raise a more important point, which is it is there's so much culture in moving to cloud. There's people, forget about how important culture is. But a lot of companies ask me, so what do I need to think about? I said, the first thing you need to think about is culture. Because if your people can't get the culture, if you can't get the culture across, then it doesn't matter what technology you're gonna put in, it's not gonna work. It reminds me, many, many years ago, I'm dating myself now. We were moving from Fork and to Sea and I had a bunch of engineers who wrote a Sea program that looked like forprint, right? You know, I had go-tos and everything, right? It was no indentation, everything. It's like, okay, well, this is Sea. The code is Sea, but it's forprint. And so you can do the same thing with cloud. And so if you're not careful, you'll end up with that. You could be cobalt. Yeah, it could be worse. And, well, I'm not sure it's worse. I have a story about that too, but. So the question, you know, the new guys is not a problem because they've come in with that mindset, they were born cloud first. And the old folks or the folks who've been doing this longer, there are those who can cross the river and those who can't, right? And then they have to embrace. If they don't wanna embrace it, you can't force them to embrace it. But there are a lot of people who really like change and they think it's great and they love it and they're into it. And there are people who say, I don't wanna deal with this. And it's okay, there's gonna be a lot. It's a long tail of legacy. They can continue working on the legacy stuff. So you really have to let people choose themselves if they wanna do that. You can't force it upon people, but there are a lot of people who've been doing that for a long time who think it's the greatest thing. Sorry, we're getting close on time here. So I wanna thank you for coming on, really appreciate it. So I wanted you to give me the last word, share with the folks out there the plan for next year. What's your goals? What are you gonna do? I know you can't talk about some of the things you're working on, because you run some top secret projects. But what's going on? What's on your agenda? Double secret probation. The state, look. It'll be on Silicon Valley HBO, don't worry about it. I'm gonna be involved in helping write this year. But look, like I told you last year, we are, whenever you're trying to understand what someone is doing, you have to look at not what they're saying, but what are they investing? What are they moving? Because that tells you, right? Where are the monies being put? Where are the senior executives are focused? That's where the action's gonna be. And you can see that we're investing very heavily in cloud. So you should assume we're going to accelerate that. We believe we have tremendous opportunity. We already have a great start with OpenStack and cloud system. And so I think what you will see is both deeper and broader. Could you be specific on some tactical goals, like version, distro, upgrades? I would look. Everybody in the OpenStack, I believe that next year at this time, we will have some very large customers running our distro. Big guys. I'm not talking about, you know, small customers love them, whatever, we'll have a lot of those. But we will have some big guys, you know, doing big things with distro. And I think we will have some of this incident, this trove, that will surprise people in terms of how it's helping enterprise on this journey. All right. Secretary of the United States President, Chief Operator of HP Cloud. Always great to see you and get the insight in the trenches, making things happen, running the operations here. This is theCUBE, extracting the signal from the noise. We'll be right back with our next guest after this short break.