 The Cube at OpenStack Summit Atlanta 2014 is brought to you by Brocade. Say goodbye to the status quo and hello to Brocade. And Red Hat. Here are your hosts, John Furrier and Stu Miniman. Okay, welcome back everyone, we're live in Atlanta for the OpenStack Summit. This is Silicon Angles, the Cube, our flagship program. We're here to go out to the events and extract the signal from the noise. I'm John Furrier, the founder of Silicon Angle. I'm Joe Myko, Stu Miniman, analyst at wikibond.org. We're here with Cube alum, Targulai, Senior Vice President, Chief Operating Officer for HP Cloud. So, welcome back. It's great to be here, as always. We did last year in Portland, you were on, and we started to see the groundswell of the HP Cloud. You were instrumental in really building out, laying down the tracks for Cloud. Now you have the big billion dollar announcement. I think I asked you at HP Discover, what are you guys going to do? Hear that shot heard around the world, and you said with a little smirk, something's coming, something's coming. So is that the shot? Is that the big aha for HP? Was that the news you were referring to? It's the beginning. It's the beginning. I mean, we have obviously, we're very serious about Cloud. We've been serious about OpenStack. About 300 people here from HP. You know, a Hylian announcement. You need to see the Hylian announcement. Obviously, there's been a lot of work to get the Hylian, but that's just the beginning. I think we should see acceleration of our investments, acceleration of what we're doing in Cloud, acceleration of our capabilities in the next 12 to 18 months. So I would say Hylian is like the beginning of something. HPU also talked to me about how HP is actively involved in the original eight formation of OpenStack. And we had Monty on Monty Taylor and Eileen Evans yesterday. I had a great chat. What's what's what's going on with you guys now? Obviously, the participation is there. We see the presence. You guys have huge presence here, both in terms of, you know, staff, engineers and all developers. Also, you're the cloud team, PTLs. What's what's the update internally? I mean, it's got good momentum. What's the internal like this cloud? You know, we've always been like, hey, what's the HP cloud? Give us the update on what's happening internally at HP. Are you guys mobilized around Cloud? What's the plans? Well, I mean, again, you always, to me, it all comes down to OK, how many people was working? How much you have it, right? So I have, you know, over 1,000 people on cloud, you know, way more than I had dedicated to this, way more than we had a few years ago. And so that's the first part. And, you know, you can see that by looking at our contributions, you know, whether it's the fact that we have four PTLs, whereas the fact that we have three people on the technical committee. And so, you know, we're just accelerating our investment. Our view in general is, you know, we and we've mentioned this before is that if OpenStack succeeds, HP succeeds. And so we are going to continue to do whatever is in our capability to make OpenStack successful. And I think now we're starting to see things are finally getting there to the point where, you know, we're talking about users and deployments and less about all kinds of architecture and things like that, which I think is really important because ultimately, the success or failure of OpenStack will be judged by its usage, not by the amount of press written about it, not by the amount of hype, but how many customers are deploying and there's lots of customers now that are starting to deploy, pilots and so on. So, you know, we're spending a lot on HP and cloud. And, you know, one of the things you can see was Helian and we talked a lot about, you know, Converged Cloud, we talked many years about that. And, you know, you need to think about Helian as we finally converged, right? That's the convergence. The work that was done in Converged Cloud has now converged into Helian. And so now we do actually have one product line, was one core technology, all that is in sort of one place and now we're going to leverage that to move even further. Yeah, it's to use the football analogy. We always like to use sports analogy. You guys had broken from the huddle. You've converged internally. You've settled on architecture. Now the team's essentially executing. What is the vision for the plan? I mean, you know, West Coast offense, if you will, but is it like multiple fronts? Is it public and then enterprise? Because Helian kind of teases out a little bit of both. You have, you know, obviously open source contribution and then you have the enterprise focus. Yeah, I mean, I think, look, first of all, we are focused on enterprise and service providers, right? We're not focused on the super cheap, you know, I want, you know, one cent storage, you know, I think public cloud is a bit of a misnomer. I think the end of the day, you know, really it's about consumer build. And you know, we are focused on giving customer solutions for both consumer and build in a hybrid context. And we believe we're just the beginning of that whole transition. And when you, and we're focused in providing solutions for that, we believe ultimately the best solutions will be based on an open source stack, whether that is open stack, whether that is cloud foundry. And so we're making huge investments in that area. And we're making huge investments with our customers to be able to help them get on that journey to the cloud. Again, both in a build context and a consume context. And you know, we have a large business across both of them, but part of, but you know, we're focused in helping our customers make that transition from sort of the scale up world to the scale out world. Sarf, can you give us a little bit of insight about what it takes to build that HP cloud? You know, talked about scale, you know, there's, I mean, a lot of effort to, you know, just build that infrastructure, maintain it, operationalize it. Talk to us a little bit about how it is to build that. And you know, how does open stack fit into it? Is all your cloud open stack or you know, what pieces are today? Sure. Well, look, we've had cloud and we've talked about this before. We've had cloud at HP for years, starting with cloud system and CSA many years ago. We've had a lot of organic activity and over the last two years, we've sort of been pivoting more and more to leveraging open stack, both from a build perspective and a consume perspective. And you know, we've broke our teeth, you know, deploying public cloud on open stack from Diablo, that was very interesting. The good news is we learned a lot and a lot of those lessons that keep on emphasizing are being put into our Helium distro. And you know, we're at different levels, but over time, right, most of our cloud efforts will be based on open stack. Today our public cloud, you can get cloud system eight, sorry, on private cloud, you can get cloud system eight with open stack today on it. And we have a lot of deployments with that. Obviously our public cloud is running open stack. As we said in Helium, we will be deploying a lot of our Helium distros on our managed cloud ES footprint, so we're gonna be doing that as well. And over time, we will be migrating a lot of the solutions that aren't based on open stack to open stack as well. But you know, look, you gotta provide customers the solutions they want. A lot of customers still have a very large part of their application footprint on scale up solutions and we're there to support them. We have one of the best scale up solutions in the world with cloud system and with CSA, but we are also gonna be helping them migrate to the scale out world. And what most customers are doing is, you know, they're keeping their scale up and continuing to invest there, but they're building a scale out infrastructure to write their new apps on. And so we're there to help them across that bridge. And part of our hybrid message is the fact that you don't have to jump. You can do both. You can have, you know, apps that some of them live in the scale up world, some of them live in the scale out and we can provide you management, instrumentation and solutions to help you manage that. So, can you walk us through a little bit, you know, over the last two years since Diablo, what have you learned? How much of that has been, you know, brought back to the open stack community and just as a cloud provider yourself, can you share with us a little bit of that journey? I don't know if we have enough time for that. Look, there's multiple parts, right? First of all, you know, scale was not really there, right? I mean, we had to do so much things just to get things running at scale. You know, when you do things at scale and performance, everything breaks one by one. You always find the weakest link. So there was a lot of weakest link activity that we had to run through. Resiliency wasn't there at all, right? I mean, in terms of monitoring, right? One of the beauties of a cloud is that, you know, there's monitoring. So when something goes down, you know it went down and you can do something about it. So instrumentation, really not there. Installation, which is not so much for public cloud but was not there at all. One of the reasons, you know, we now have come up with a distro based on upstream is because we finally managed to get the proper installation technologies part of triple O which, you know, Robert Collins from our group is driving. And so that wasn't there as well. So a lot of things that are sort of like, things that are more mature, I mean, it's normal. It's a maturity issue but a lot of things where mature technologies already went through that. We're just not there in open stack. On the other hand, open stack is very flexible so it has capabilities. We're still working through Neutron. I mean, Neutron still has issues, you know, as long as you're doing flat networking, everything's great, but you want to do more sophisticated things. We got some work to do. Yeah, I'm just curious on the operational side, what one of the stats, you know, we've quoted and many people have is, you know, Facebook can manage, you know, 10 to 20,000 nodes with a single administrator, typical enterprise somewhere between, you know, if they're good three to 500, you know, what do you see from both the public and private offerings that HP offers? What can we do? Well, the first thing I think that's important to understand is Facebook runs one app. Okay. Agreed. They have a very homogeneous solution. It's very different. I think we should be able to get 2,000s depending on the complexity of the apps you're running and the diversity of the base. But I think the area that we're focused on right now is the ability to just install it, run it, and without having to babysit. And on our public cloud, you know, we run very, very large thousands of nodes, but I also have a staff of people that is watching over it and taking care of it. And I think there's more work to be done in OpenStack. One of the reasons, for example, that we released our community edition now, Opelion, and our more scalable commercial edition is a few months out, is because even though we finally got triple O working and everything to get it done at scale and reliability that enterprise expect, we needed some more soak time because after all, Icehouse only came out three weeks ago. And so that's the kind of things that we're focused on. But I have no concerns that OpenStack will get there. But really now finally, I think as people are doing more things with OpenStack, we're finally pivoting to the level where there is a focus on those key things that are actually about what is it to operate, right? Forget about, okay, I can run a cloud, congratulations. What is it to operate at scale? What is it to operate? It was a small staff without an army of developers, right? And success will be when we have enterprises that don't have an army of developers adopting OpenStack. And, you know, convince that we will see that in the next 12 months. Sorry, people are talking about like the big vendors now in here and the big debate yesterday we had on CrowdChat was, can they grow organically as a community or do you need a leader to come in there? It's almost 50-50 down there and the most ecosystem need leaders and I'll see IBM, HP, huge players, Red Hat are in there. I mean, it's some big money in there, big muscle, big contribution. So everyone's always worried about is HP going to fork OpenStack? And that's kind of not worried, but that's on their mind and people have been addressing it. So what's your take on that? How do you talk to the folks out there and say, what is HP's intentions? Is that something that always gets kicked around? So why don't you address that comment? Yeah, so I think, look, if you have to judge us by our actions and our actions are to put more and more people contributing. I mean, we have four or five PTLs. We have a lot of core folks on the technical committee. So we believe that the right way to move OpenStack forward is to push the OpenStack within the system, not from outside. And so we don't, forking would be a failure of OpenStack. I think that the key is to have the people who are driving OpenStack be focused on what customers need and not be focused on interesting ideas that are just experimentation. And so I think you heard some of that also from our friends at Rackspace and so on. I think as more and more customers start deploying, I think the natural gravity will push people to focus on the things that customers care about. Before, when there were less customers deploying, we could all have nice academic and opinionated discussions about what's important. But we actually have customers, they actually know what's important because they're using it. And so we actually feel good about OpenStack and last few releases that we do think that it's starting to coalesce around what are the important things for OpenStack to be successful. And it's, again, it's just based on customers. The more customers, the more people who are using OpenStack that I see here, the happier I am, because that will drive it to a good, healthy ecosystem. And so far, we think that is the right way to do it. And we have no intentions of any forking. Our intention is to work within the community part of what's we built to make that happen. Eileen mentioned some of the governance things yesterday. It's pretty smooth there and good harmony, so to speak, good balance. I mean, there's always conflict at some level, but there's still some good trust there. Well, I mean, look, it's a very active ecosystem. If you don't want any conflict or any discussions, that can give you a dead ecosystem with no check-ins and nobody cares. I'll take some of the issues in OpenStack, some of the healthy debates, because I'm getting innovation for it. So I'll take it. I mean, we got to be on top of it, but you know, it's healthy to have that as long as in the end, you get a better solution. And again, you're focused on the customer use case, not on some archaic technical argument. Yeah, let's talk about that. I mean, IBM was in the same position. And their view is similar to yours, Higgs. If OpenStack's successful, they win, you win. So that's clear. Let's talk about customers. One of the things that we're always looking for here, we don't see a lot of it because either they're quiet, don't want to disclose it, but there's just not a lot of customer discussion stuff. I hear all-way conversations coming down the escalator, POC and so and so. So what are you guys seeing from your customers? What are some of the activity levels? If you can name names, that's great. If you can't, then just talk about some of the use cases and some of the environments that you're executing in. Sure, I mean, I'll talk to you at a broader level. You know, last year, there was, you know, two years ago, people thought, what is this OpenStack thing? Is it the right thing, the wrong thing? Then last year, people started saying, you know, maybe I should do a POC. This year, I don't have enough capacity to support all the POCs if you want to do. And you said, well, what are they doing? Okay, so the classic use case on an enterprise is, you know, they have their traditional apps and they've cloudified them to some extent. You know, VMware, scale up stuff was dedicated hardware. They've got self-service, but they know that to really achieve, you know, the scalability and the velocity that cloud perfumes, they're going to have to do, you know, a scale out solution. And so most of the customers that I talked to, and we're talking, you know, Fortune 50 companies, big banks, very large companies, they're basically looking at building a scale out cloud in-house and then starting to develop their next apps, their mobile apps, their next generation apps on top of that cloud. And what you're seeing right now is that they're doing POCs, RFPs, pilots around that. You won't hear the big stories yet because I think we're in the middle of it, but I'm sure next year we will have some bigger stories. The other part of it is some of this is tied to the velocity of developing these new apps, right? Because again, these are clouds that are very different than the clouds that are existing apps are on. And as we all know, you can't just move apps as they is. And so you've got to develop new apps or at least develop components of new apps that will then tie back through a hybrid mechanism to their old apps. But I'm seeing a lot of that, a ton of that. And I think those will be the things that are going to be in production next year. And even then, though, it's going to take some time. It's an adoption curve. Again, it's tied to the app. We were talking yesterday about, is it a product or a platform? And it really is a platform. There's so many use cases to innovate around as long as a core stable community seems to be going well. So with that in mind, what are some of the verticals that you see adopting this now? Obviously we're talking to the Red Hat guys and they're saying they're seeing financial services. It's now cool to bring open source into the enterprise. So what areas do you verticals do you see kicking the tires and rolling out some heavy duty open stack? I see it across the board, but obviously financials because they have a lot of workloads and also they're always looking to improve their speed. I see e-commerce, a lot of e-commerce, government, a lot of government, government loves open source, especially people who like to look at the source. Big transportation companies. I don't, it's pretty much across the board. And again, a lot of the applications, some of them have very specific applications, but many of them it's part of their overall view of their transformation of their IT. And this is what people need to understand. It's not about, oh, I heard about open stacks. I'm going to deploy open stack. It's no, I want to transform my current system to a more cloud oriented scale out system. I want to get that level. I want to get those efficiencies. I want to get that velocity. And now I have to make a choice of a cloud platform. And so I've chosen open stack. It's not on, it doesn't stand on its own. It's part of an overall transformation. But that's also why these things take longer. Cause these transformations don't happen overnight. It's not like deploying a new app. So whole transformation. But I can see again, financials, transportation, e-commerce, people who are saying, you know what, we need to transfer that. We need that velocity. How can you help us? And yeah, we've decided open stack is the right platform just based on looking around. And now how do we go about doing that? Sir, I want to get your take on something. If you look at hybrid environments, one of the challenges that we have is even if I have similar infrastructure or even similar virtualization in my environment, the customer owns and where I go to, the way they're configured tend to be pretty different. How important is it in your mind that we get, I don't know, it's not homogenous, but you know, the kind of same building blocks or same architectures in both sides of whether it's my environment or you know, my service provider or cloud provider. Sir, well first of all, I think there's misconceptions about hybrid, right? Hybrid means many things, right? The very basic view hybrid is, oh, I can move my workloads around, right? Turns out that's not the prevalent use case. The prevalent use case is, I want to build a composite app with a system of engagement in one place and a system of record in that place. In that case, it doesn't, they don't have to be homogenous. It's gonna be very heterogeneous. In fact, you can have a scale out model connected to a scale out. That's how a lot of people build mobile apps. In the case where you do want to burst, was it DM? Or federated applications, right? Yeah, which is common. You have to have certain layers that are the same. It doesn't have to be pure, but you have to understand your service layers and your interactions with your APIs and there you do have to have some consistency. And you know, I think we're getting there. It's not perfect. I think it's still the point where you have to actually look at it before and there's actions in the world. So I think we're getting there, but I don't think you need exactly the same architecture. I think, you know, the whole idea of cloud is to abstract and so that would be, you know, going counter to what cloud's all about. But, you know, like anything, you need some API consistency. So I asked the question that the healing announcement then get the answer on CrowdChat. I was looking for, so I'll ask you here, how are you going to spend the billion dollars in terms of, is that code? Is that for coders? Is that for evangelism? Is it for outreach? Is it for staff? Can you kind of give us a high level kind of breakdown of you guys going to deploy that funds? So I can tell you at a high level, there's a few things that will then let you think about it. I'll give a connect to the dots. You can connect the dots. First of all, HP is a very conservative, HP lawyers and finance are very, very conservative. So, you know, we're probably spending a ton more, but we could only state this because it could be, you know, deduced in this way. But we're spending way more if you look at our overall thing, including data centers and so on. So when we talk about this, you should be thinking and you look at our investment and how many people we have contributing to OpenStack, how much we're in the market, how much we're doing that with Cloud Foundry. You should assume we are dramatically increasing our investment in the R&D aspect of things. I'm not going to tell you that's where the billion is. I'm just saying, you know, people like said, oh, well, this is very small if you think data centers, they're right. But again, HP is very conservative of what we can report. So in general, right, we let you see how much we're hiring. You should think there's a big R in that there's a big R and D in that number. I can't get into the details, but there's a big R and D that number because in the end of the day, if you want to make open source successful, you have to participate and be part of the community. And we're investing, you know, very, very heavily there. So it's not a big bag of money for the foundation. So here's a billion dollar donation. It's really more of your investment, HP's investment into the efforts of OpenStack and the community. Correct. Like, you know, one of the things we said I believe was said during Hylian, but if not as like, we're going to focus a lot on upstream, right? One of the reasons we, the way we're, you know, the difference between CloudOS, which was done before some of the maturity in OpenStack to what we have today with Hylian is that Hylian by definition is defaulting to upstream. Okay, that requires a lot of people to do that, right? And so definitely, you know, you're going to see, you know, I think we're number three or number two, I think an ice house, expect that number to rise, expect us to have more reviewers. Reviewers are actually more important than code contributors. You know, we already have three, four, five PTLs expect us to have more. And again, it's not because we want to dominate, it's because we have an agenda. We want to make OpenStack mature. We want to be able to get OpenStack to the point where enterprise don't think twice before adopting it. So, Sar, you know, everybody knows HP, you know, leader and infrastructure, you know, took server, storage, networking, you know, towards the tops of all three of those. Can you speak to the development of kind of the software team that you have there? You've got a thousand people in Cloud, you're contributing more to what's OpenStack, you know, out of a thousand, I mean, is that 90% software people? Or, you know, where does kind of the software mojo fit into HP? Okay, so first of all, we have a large, large software organization developing software tools and orchestration, which is separate from cloud, but part was cloud, things like cloud service automation and all our other platforms. So, HP's got a very large software business on its own, which is unrelated specifically to cloud. In the cloud space, we have, we don't break out the numbers, but you know, I would say that I have probably twice as many engineers working on cloud today as they did a year ago, and I'm not stopping. So, you should assume that, you know, you know, we are sort of the open source destination for someone who wants to work on open source on cloud, you know, we are a destination company because you get to work on the latest stuff with support and you get to be involved in the latest technologies, whether it's OpenStack or Cloud Foundry and so on, and you're going to see increasingly our participation there just to continue increasing. So, I can't break out the numbers to you, but I would just say that we grew significantly last year and we're on the same trajectory. Sark, talk about the operations compared to where they were last year, I see a thousand people working for you, you're an ops guy, you'd like to run the trains over there, make them on time for everything else. And you got Martin Fink involved, now he's an open source guy, talk about the balance between your operational agenda and your plan versus the, I call it the, not the R&D, nothing's an R&D guy, but like it's an open source kind of ethos that Fink has. And how is that culture being bled into the operations? Because, you know, HP has a cadence, you know, good customer satisfaction, but it's HP specific, now you have this open source ethos coming into the culture. Talk about that blending, how does that, how's that going? And then give us a view inside HP. Sure, so first of all, over the last year, we have, you know, helped, we brought in some big hitters in terms of open sourcing cloud like Bill Hilf, right, who did a lot of stuff in Azure, but before that was a big open source guy, we brought in under people and so on. And then on top of all that, right, we moved under Martin Fink, who has always been, quote unquote, an open source guy. And that's been great because, you know, we were sort of, we are sort of driving open source not only in terms of this, but also in terms of how it drives into the organization, into HP in terms of the whole culture of, hey, let's open source this, let's do it as part of the community as opposed to, we'll sit in some dark room and do it. So I think, you know, Martin has pushed a lot on that. I think Martin, when he came in, you know, initially said, let's make sure everything is upstream, let's make sure push, you know, we push all that in that way. Now, you know, from an operational standpoint, we're still running a company, but we are separate from HPs, sort of, you know, HPs divide into different pieces. We have our own cadence. We have our own release cycle. Part of the move to open source is also doing CICD, like we talked about when we did Healian, we're gonna release something every four or six weeks. So I think it's blended well in terms of that. I think there was latent open source stuff going on in HP anyways, but I think now there's sort of a thing on it that says, okay, yeah, this is right. This is the way to do it. You know, I lean, did a lot of pioneering work on how to drive open source. So I think it's all just now come together before it was sort of in different pieces. She was awesome yesterday in talking about the Linux board and her role also here. I then asked the indemnification question. I'm obviously, that's a huge deal. You guys have some history there. Talk us through how that went down internally and why is this indemnification guarantee, I guess, like a better word, important for customers? Look, we, our view is we want to make it as frictionless as possible for people to adopt open stack. And, you know, we have a history, right, from indemnifying Linux back in the day when people were trying to stop Linux from adopting, and so we looked at this and said, as we thought about, as we said about going to Helion, we said, what are the things we can remove? What are every possible barrier we can remove for people to think twice before using open source? And while open source is very popular, there are still people when you talk to customers, they're like, well, what about this? What about that? Let's take it off the table. So you still as an sales inhibitor, basically. Yeah, it made people think twice, not everywhere, but we definitely had a lot of customers come to us and say, and because we have a history of indemnification, we said, hey, you indemnified Linux, why won't you indemnify this? So we said, let's just take it off the table. How does that work? Is it like legal documents? Is it the dual source? Is it the source code that, what's the specific value of indemnifying HP's products relative to the open stack? Is it certification involved? If you, again, there's a legal document that explains it, and I'm not gonna, I'm not, you know, you should have asked that. But basically, from a deep risk. The basically point is if you use a licensed version of HP open stack and you sign the, in the case of an agreement, separate agreement, you have to sign, then if a third party sues you because of violations relating to open source, proprietary, so forth, that specifically not, I mean, if you use it for something else, it's your business, then we will indemnify you. The same way that those, you know, jokers from SEO were suing people for using Linux, the same idea. It kind of, it freezes the market, basically. So you essentially take that de-risk, take the de-risk. Take it off the table. Don't worry about it. If they come after you, HP's here, we're a big company, 100 billion dollars, we'll take care of it. You can go focus on your business. Talk about the open stack progress so far. I mean, obviously it doubled since Portland, almost 4,200 people. Some thought there might be more here. A lot of activity. What's your take on the evolution of open stack? Early days still, you know, National Anthem hasn't been sung as, are we in an inning? I mean, come on, tell us your take on that. We've made progress. Again, I think it's grown up. I think we're way out of the, we're probably in the fourth inning, maybe. The difference, I think, again, the question is where's the gravity of open stack, right? I would say that the gravity was always been with, oh, this is cool. This is the developers. Look what I can do. The gravity now is moving more and more to, okay, I'm a customer. I'm making a strategic choice. Make sure that this is a good choice. And so that's where the gravity is moving to. But I think this is an important time for open stack over the next 12 to 24 months. I wanna see more and more users in these conferences, and especially speaking, not only attending, but speaking and talking about what's important. Like we had AT&T talk about it, users. That's sort of the transformation, right? When you look at the curve, you start off as this really, really cool technology. Everybody's excited. And then, oh, we need to use this. Okay, so there's all these other things we didn't think about. So I think we're there. I mean, with Ice House and now with Juno, we're getting there. Ice House, we finally have installation. We have better on upgrades. There's some instrumentation. Things have gotten better. But we're now at the point of, okay, now this is great technology. How do I make it usable for mere mortals? Talk about the CIO. I told the CIO last year in the like, oh, I love open stack. I'm like, do you know what open stack is? Oh, no, I just love it because I don't want to use Amazon. And that was a legit answer, you know? So that was last year. So this year you're starting to see a little bit more answer, more coherent around. I want to look under the hood. What are you hearing from CIOs? Why are they so into open stack? Is there a particular affinity towards anti-Amazon? Or is it more of I just want to have some legacy integration and some compliance issues? What is the rain driver for open stack success? Or I should say the demand to do more. Yeah, so like I said, I think we shouldn't be looking at it from open stack out. We should be looking at it from their initiatives in, right? So the CIO under pressure to move faster, you know, drive faster innovation, have more flexibility, deploy applications faster, all these beautiful things. So then you look around and you say, okay, how do I do that, right? So, you know, there's Amazon or those guys, but then they have their dynamics. First of all, public cloud, as we know, if you're a big company, it's very, very expensive. Second of all, it's got its own challenges. And most of these have hybrid. So he says, okay, how do I do a private cloud? He looks at private cloud, he said, okay, what's available in private cloud? And then he hears about this thing. Okay, so there's all kinds of proprietary stacks with people trying to get away from proprietary. And he says, oh, there's this open stack thing. What is that? And he says, okay, well, you know, that's the most popular non-proprietary solution. And that's how he gets to start talking about open stack. And then they start digging and saying, okay, well, who else is deploying? Who are the vendors I can talk to? How do I get educated and so on? That's typically that path. Now, before it was like, what is open stack? Now it's like, is it ready for prime time? So we've improved. Next year, I would like it to be, okay, what's the biggest installation? And you know, when can I deploy? Yeah, and some groups swing on some use cases. Exactly. Okay, final question for you. Share it to the audience in your own words. Why is this year so important in the industry and open stack? Why is all the buzz around all these forces coming together and tectonic shifts? I see cloud mobile and socials on everyone's mind. Well, what's really happening in your own words? Well, I don't think I'm saying anything new to people that the world is moving to cloud. And a lot of cloud will be private or hybrid. That's a fact. And open stack looks like it has the opportunity and the great opportunity to be the technology to do that. And so that's amazing. I mean, if you think about back a few years ago, you know, we have everything set up for our success. We just have to execute. Okay, we're here inside the cube at the open stack summit. Sargillai, senior vice president, chief operating officer at HP cloud. Thanks for joining us. We'll be right back with our next guest after the short break. Thank you. Great to be here.