 Live from Mountain View, California, it's The Cube at OpenStack Silicon Valley, brought to you by headline sponsor, Mirantis. Here are your hosts, John Furrier and Jeff Frick. Okay, welcome back and we are here live in Silicon Valley for the OpenStack SB event, OpenStack Silicon Valley. I'm John Furrier, the founder of Silicon Angle. I'm Joe, my co-host Jeff Frick, general manager at Cube Silicon Valley operation. Our next guest is Gary Chen, research manager of cloud and virtualization, system software at IDC, international data corp research firm. Welcome to The Cube. Yeah, thank you guys for having me on. Great to have you on. I'd love to get the analysts on perspective because you want to get, you got, you track it, you're tracking this market. So, you know, we were talking prior on our intro, what's the TAM on this thing? We don't know. I mean, is it cloud, green field, OpenStack sales or is it data center disruption? So it's kind of not clear. I mean, certainly the data center players are here defending their, you know, HP, you know, IBM's not here, but that's where this is going. Software, we had Mesosphere on earlier, classic collision to the data center, or is it cloud and another TAM? I mean, make sense of it for us. Do you have any insight there? Yeah, the, you know, it is, I mean, I think cloud is like, you know, it's, it's, you know, this new idea, but it uses a lot of existing technologies, right? So a lot of, you know, the existing stuff, you know, hey, you know, we'll put it and make it a part of the cloud because the cloud still needs networks and storage. You know, some of it may be slightly different in a cloud, but then there are like new markets that are created. So, you know, the kind of stuff I look at cloud system software, a lot of what OpenStack and that, I mean, that doesn't really have like a clear category to put it in. So, but the way it's sold makes it, may make it difficult to extract, you know, this is a market and this is a TAM, right? Because a lot of people are like, well, we're selling this, you know, bundle of stuff, which is, you know, maybe it was a bundle of virtualization, then we had some cloud stuff or someone has like an operating system and, you know, they add OpenStack as, you know, basically a package that you install on an operating system. So, yeah, I mean, I think- And there's new approaches, you mentioned that as a service, it's interesting, right? So it's not clear, is it? Yeah, yeah, it's not clear at all. And, you know, so I mean, we are trying to, you know, start to make sense of that. We have some, you know, tixonomies and we're trying to, you know, gather, you know, information on revenue, decide how to split it up and do a forecast. But, you know, I think right now, you know, any way you put it, if you look at kind of OpenStack, whatever the market you want to call it, you know, the market is still very, very small. And it's tough to know where to put it yet. It's a new creature, if you will, in terms of categorical definitions, not like a nice little tight package yet. Exactly. Okay, so that being said, the Software Defined Data Center, we were at VMworld this year and clearly Docker stole the show, containerization. We heard, again, Mason's here talking about scaling at scale, the containers, the app developers are in charge. So the impact to that on the data center guys, you got networking compute storage. What are those guys going to do? How are they responding to this disruption in this converged infrastructure, which is going to end up becoming some sort of commodity. I mean, Amazon is driving to cost a zero while innovating. So they're adding more features and lowering their price. That is certainly a direct target at the storage compute network guys. What's your take on that? What's your commentary here? Yeah, I mean, I think a lot of what's happening there is people want to make that more of a kind of a more malleable resource, right? So those things are still essential, but I think how that's being consumed is totally changing, right? So if you're managing those things in enterprise now, I think today someone probably comes up to you and says, I need this storage, I need this server or something like that. I think a lot of that's going to change. It seems to be going through all of these portals and things out in self-service, different layers of how you can service a customer. So I think those guys will be charged with running, that's basically this infrastructure as a service and they're going to keep that running. And they'll probably fairly insulated from what the actual end user is buying. And so there'll be layers of paths and web portals and things like that kind of between them. So it does become more widespread and more available, but it also does become more abstracted. And so I think for the market, a lot of the value is going to be in a software layer that's above that. You know, I got to ask you, obviously the breaking news this morning, we've been reporting on SiliconANGLE as Docker gets a series C for 40 million. I heard about that. I mean, Jerry just financed the thing in less than a year. They already got, they're in their series C round. I mean, that is the ultimate hype machine in action. Right now they have great traction, so product market fit seems to be there. Okay, how does that overlay into this market? Funding is an issue because a lot of the companies here in the OpenStack community are about three years old. They're either going to tank out or get another round or some sort of consolidation. How do you view that expansionary funding versus the consolidation? Yeah, so I think right now in OpenStack, I mean, everybody is doing a distro of OpenStack, right? The market can't possibly support all those things. Even OpenStack is this runaway success, which very likely could be, but there's certainly going to be a lot of consolidation there. And I think the thing is it's really hard to differentiate an OpenStack right now, at least kind of on the software distro side. And so a lot of it may really just come out to, if you can just dig out, if you can manage to find some kind of niche, some small market. So if you get some customers there, or you build some expertise there, or it may just be for people, little pieces of technology, I think that one thing that is good for, a lot of these distributions is that the scope of OpenStack is expanding. So there's a lot and a lot more modules. And I think it really starts to let people specialize. In the beginning where it was just kind of a core set of modules and everyone's OpenStack included these and it wasn't much to do, but if you really want to focus on Paz, maybe that's where you focus on, or Trove at databases, a service if databases are specialty or you have install base there. So I do think we will see more differentiation, but I think the consolidation is going to come. But when that's going to happen, it's hard to say because I mean, in some sense the OpenStack market hasn't even really has just started, right? So it's not like there's lots of customers out there. So the consolidation still could be, we still may be in that expansion phase. Not a lot of net new customers. I mean, there might be a little bit of new business, but it's the disruption piece really that you're watching, right? That's kind of what you're saying? Yeah. I mean, I think for right now it's people are looking at OpenStack as an interesting new technology. It's probably not quite there yet in terms of mainstream adoption. So I think we'll probably, right now we're seeing, I think more startups around OpenStack form at this point. So it still seems to be expanding, but yeah, those disorders will have to collapse at some point. So Gary, let's shift gears a little bit. You had a panel earlier today, I believe. Yes. So talk a little bit about what happened on your panel that you had on some of the interesting insights that came out of... Yeah, so the panel was called, it's open source, winner take all game, right? So we had Martin Mikos from Eucalyptus. We had Steve Wilson who was from Citrix representing CloudStack, and then Boris Rensky from Morantis who represented OpenStack. So the whole point was kind of just to discuss, is there kind of supposed to be just one winner, right? At least on the open source side. All these platforms battle it out and then there's one left and that's kind of the open source option and then you compete against all the other proprietary like Amazon and all these other guys. And so, I mean, I don't know if there were any really solid conclusions that came out of it. I think any time you get those few guys together it's kind of really competitive which it was and it was really fun. But I think in the past, the lines seem to be more clear. I think when Linux, like the other big open source project was coming up, people spun it as kind of open source versus proprietary. A lot of it was like, can open source work? This is model work. Today, yeah, I mean, I think we know that it does. And then it was also kind of like, there were clear villains, right? It was Windows. And to a lesser extent, it was Unix. At least that's kind of how the market and press and media kind of spun it as. And I think today it's, yeah, there still is some of that open versus proprietary, but there's a lot of blurring between some of these things, right? So you have Amazon, but they don't sell software, right? They're only a service. And so that's different. And then you have these platforms that compete at a lot of levels, but like Citrix has a lot of assets that it makes work with OpenStack, with Zen Server and NetScaler and all these things. So they're heavily involved and they're contributing code. And so it's probably not just this versus that. And I think most of the participants were still pretty adamant that their platform was still out there and still running the race. But I think it has become, I think, a little bit more collaboration. Before it was, I think, much more like us versus them. But I think Citrix is really focused on desktop and app virtualization. They'll try and make that run on any infrastructure, whether it's CloudStack or not. And of course, Eucalyptus got bought by HP. So there's certainly going to be some kind of eucalyptus, OpenStack kind of love, exactly how that's looked no one really knows. But yeah, so even Martin put on an OpenStack shirt during the thing, so he was wearing a T-shirt that Boris gave him. So that might have been a sign that something's changing that he actually put that on. But at the end of the day, it's about market traction with customers and market traction with the developers. So I mean, are there any kind of early indicators that you see or you're tracking to know potentially who's kind of gaining momentum and maybe separating themselves from the fact a little bit? Yeah, I mean, I think when you look at production deployments today, it's so small and it's really kind of the cutting edge of users. So it's hard to kind of make assumptions from that. The way I kind of look at it is kind of right now, I think the best kind of measure of where it's going is kind of developer momentum. So I mean, there's a finite number of developers out there, right? So you want them to work on your project, right? So, because that means something when it filters downstream into these commercial products. So the more developers you have means the faster you can innovate, right? The better quality yourself. We've got more people to test it and things like that. So I think when you look at developer momentum, that I think that's a good indication of the project. Because I think if you look at, people do feature comparisons today of where things are. And that's pretty interesting if you have to deploy right now, but not everyone has to deploy right now. So I think long term for open source projects, you really look at the community of where that momentum is, and kind of any shortcomings eventually get solved once you kind of build that momentum and you get those resources, you can overcome all those things and that can happen very quickly. You can catch up and you can build out your stuff. So I think right now the developer momentum is kind of the interesting thing to see where the industry is putting its developers. And most of it is coming from vendors, right? Vendors support open source, so they have to strategically kind of allocate where they want to go. So the open stack is definitely in the lead there. Gary, we've got a minute left. I'm going to get your take on the HP eucalyptus deal. Obviously you said you're on a panel with Martin. We've known him for years. Great guy. Open source guy. So it's Martin Fink. So what do you think that's going to do? Is that a yard sale? Was it strategic? So he's a big power player, face of HP, potentially the face of open stack. What's your take on the deal, pros, cons? Yeah, so I mean, I think there's a lot we don't know, but what I would suspect is one, it was partly a talent acquisition, right? So Martin is really well known. That team has experienced an open source. It doesn't mean that they have an experience with open stack, but just with open source models in general, right? So HP is really serious. It needs more people that really understand open source, how to foster that. So that was probably part of that. The other part is the Amazon thing, right? So eucalyptus was really the only vendor that had kind of this API license, right? So, but it's not clear like, if HP is going to inherit that or not, right? So... Or embrace it even if they inherit it. Yeah, it's kind of unknown, right? Yeah, or even if they want that. That could be a potential asset that could be valuable. I asked them a direct question. How do you keep up with that moving train called Amazon? I mean, Amazon's adding new features at a clip. We're seeing amazing traction and the pace of new stuff there, features. I mean, last year was some game-changing stuff, Kinesis, Redshift, doing well. Yeah, they're adding a lot of things, and I think Amazon isn't really clear like what they consider their APIs really something that are going to vigorously defend. They seem to be letting people use it now and not really caring, but so kind of how they react and what HP's kind of plans are, if they really want to... Well, any chance he's not a new to the business, he's a strategic thinker and they have a mission. So I think that'll be a pretty easy decision for them to help their mission or hurt their mission. That's going to be the strategic question. How do they play the competitive strategy game on that? Yeah, yeah. I mean, I think the big thing for HP now is like they basically have two platforms that do the same thing, right? They have competing platforms. So they have to figure out what they're going to do. Are they going to try to sell both of them and kind of spin one as this one is for this and this is for this? Or do they kill one off? I don't think merging is possible. I just don't think you can take a product and this product and all of a sudden magically make it merge. It may sound like it's... It may sound easy to say that, oh, we're going to merge things. Well, you've got internal five terms within HP, but the issue is you can't be half pregnant as the expression goes, right? So HP's got that issue. The other thing that's interesting I want to get your thoughts on along those lines is as a portfolio, HP has a pretty big portfolio within the cloud itself. It's pretty diverse and different. Tim Cook was on Charlie Rose's Friday, he talked about Apple, how they simplified and he said I could lay all my products out on this desk right here and that's our ex billions of dollars a quarter in revenue. Okay, so basically they would simplify their products. Is that even possible for an HP? I mean, and is that something that they should do? I mean, I think they should. So even if you look at just cloud platforms, I mean, right now, I know of at least three they're juggling, right? They had the older cloud matrix platform. Now they moved to OpenStack and now they've acquired eucalyptus, right? So, yeah, I mean, I think, and those things kind of is at the end of the day, they do kind of the same thing, right? So I think it's very confusing for customers, and you try and make it really specific to that or that. It just gets really, really complicated. So yeah, I think it's easy to say that yes, they should simplify, but they just have this massive, I mean, they're a massive company. They got these customers that are probably about long-term contracts into things like that. So yeah, I think it's really, really hard to do that, but I do think they should progress and simplify. Well, the first thing that Martin said he would do is talk about the people issue, go look at the people, get to know the organizations, and that includes the sales force, right? You got to know who's touching the customer, right? And then the product market fit from there. Gary, we got to get the hook here. We got to go. Thanks for coming on. Thank you for taking me at the analyst perspective. Gary Chen from IDC breaking it down here, covers system software. That's the operating system of the future. That is the cloud, hybrid and public. Now this is theCUBE. We're sharing all this data with you. We'll be right back after this short break.