 Computer Museum in the heart of Silicon Valley, extracting the signal from the noise. It's theCUBE, covering OpenStack Silicon Valley 2015, brought to you by Morantis. Now your hosts, John Furrier and Jeff Frick. Okay, welcome back everyone. We are here live in Silicon Valley. This is Silicon Angle Media's theCUBE, our flagship program. We go out to the events and extract the signal from the noise. I'm John Furrier, my co-host Jeff Frick. Here at the Computer History Museum, we have for the first time on theCUBE, as an IBMer, Monty Taylor, who's now Director of OpenStack Innovation at IBM, formerly of HP, been on the board of directors for the OpenStack Foundation, technologists, great guy, CUBE alumni, again, formerly of HP. Welcome to theCUBE as an IBMer. And whilst we have all the IBM events, insight coming up, and obviously we do a lot of coverage of IBM. Welcome to theCUBE. Thank you. Thank you very much. Okay, so first question, why the change? IBM poached you, you solicited, did you leave on your own? What's the deal? So, they're going to love this answer, but so all things start with going out for drinks with Jesse Proudman in New York City. We should, yes. And really from there everything is history. And so Jesse and I got to talk, and we chatted a few times before because he's an awesome guy, and he started letting me know that OpenStack, that IBM is going to be doing an OpenStack public cloud, like Amazon scale. They were going to go after the big guys, right? And actually do it. And that's a big proposition. That's a really big deal to do. And it's a thing that I think is important for OpenStack, important for all of the community. And I want to be involved with that. I want to make sure that- Is it a moonshot or is it realistic? I mean, you're my host, you're going for it, right? It's totally realistic. I mean, think about the companies that could do that, right? Like this isn't something, you're going to go get some VC to go chase that, right? This is the big players that already have the metal on the floor. So if you look at the software acquisition, we got a whole bunch of data centers out of that. Already had a whole bunch of data centers. In fact, I'm pretty sure they got data centers in more locations than Amazon does, right? So once you get the hang of this, you got global reach, you've got all of those things. And you've got a company with the history and the heft to go and do something big. Like not to do it just a little bit, but to actually- And you got the announcement from IBM about getting 90 days blue box down to software 90 days, pretty big deal. That's exactly showing exactly how serious all this. We've got the software property, we've got that. And now we're delivering OpenSec into that for managed private. So even though I'm a bullish fan of IBM strategy, what they're doing, I got to put the skeptic hat on. Sure. Amazon has a history decade of trajectory in the economy of scale. If you try too fast to rush it, there's diseconomies of scale. Are you guys have a blind spot there? Do you see that? Are you aware of it? What are you guys doing to address that? I mean, so there's a few things there. First of all, they do have 10 years of experience at scale, but I'd argue that IBM has over a hundred years of experience in technological innovation and they've been doing this cloud thing in a thing called mainframes for much longer than Amazon has been a twinkle in anybody's eye. So that being said, you're exactly right. It's a big thing. They started off as an online services company and know how to run services architecture and know how to do those sorts of things. And that's not a thing that you should chase lightly. So there's a few different things. And the company's super behind it. We heard Ginny's been talking publicly, we're all in on cloud. Yeah, they're all in on cloud. Apacheano comes on the queues, like we're all in on all this. Yeah, totally driving that, totally driving that hard. It's a thing that they're into, but there's sort of a couple of things. One, bringing in me and some other folks, we just hired Jonas Chakobi, who's super exciting. I'm going to be working with him. So we're putting really smart people on a lot of the hard problems. And actually, this is what we're talking about just a little bit ago with the room of analysts. One of the reasons I'm excited about this for the OpenSat community is that we're going to be able to have a tight feedback loop. Me and the folks who are going to be on my team, we have history of knowing how to work inside the OpenSat community really well. So as we find the things that are the problems at 10,000, 100,000, a million nodes, we're going to work very well with the OpenSat community on solving those. Oh, and IBM, it may not get as much credit as it should for the online properties that it does actually run. Like it's- Well, and also their history in open source. I mean, they took a bet on Linux when it was a second class citizen. Exactly. When IBM had a lot to lose betting on open source at the time. That's exactly right. And in fact, I'm going to get this wrong because I've only just did my IBM orientation yesterday. So I'm going to get all the dates and the tales wrong about this. But IBM's actually bet the entire company on entirely new technology before in its history. You mentioned Linux. Back in the, I think it was the 30s. Hang on, yeah, 30s was the depression, right? See, this is- Oh man, I'm in trouble. In the 30s, everybody else was having financial problems. IBM decided to double down and put all of its money into the tabulating machines for tabulating things. Right after that, coming out of the depression, the government had a giant need for number crunching. And the only company that was built up enough to be able to take to work was IBM. So they basically during the depression, doubled down bet all of their money on an entirely new technology front. And it kind of paid off for them in giant dividends. So this isn't a new thing for the company, they've done it before. Well, we've been covering IBM for a couple of years with theCUBE and we saw the initial announcement and you can see the messaging was tight. They've been executing on that. Bluemix is out there. They've still got some work to do. Amazon's got Beanstalk, Bluemix, I don't know if they have yet, but they'll get something going. It's clear that IBM wants to meet Amazon in the middle of the field. The question is whose terms is the battle going to be on? That's us to speculate with the media company. I think that's actually, I think that's exactly right on. And that's one of the things that I, I think I've said it here before on theCUBE. I want to make sure that what OpenStack, that OpenStack isn't playing the game of chasing Amazon because there's a lot of ways in which I don't think that Amazon's designed their thing in the way that I want as a user. Like they've got some, don't get me wrong. There's some great, obviously, you know, they're doing a great job in a lot of ways. But I want the cloud to serve my needs as a user. I don't want the cloud to be an emanation of what my already existing data structure, infrastructure needs allow me to produce, right? And this is, Amazon's design is there because Amazon has a particular economy of scale. And so what you get out of them product-wise is the thing that makes the most sense for them maximizing the investment of stuff that they've got. From the OpenStack perspective, I think we can do more than that. I think we can serve a wider range of users. And I think that when we do serve a wider range of users, and we actually are in service of what users need out of us, then I think that that's us setting the terms of the argument and putting things on the table that make absolutely no sense for Amazon to chase, right? Like things like the OpenStack's ability to have private cloud, you think about what's going on with BlueBox and actually getting on-demand private clouds in the same data centers that we're going to be rolling out, you know, a global public cloud in. Let me just think about the opportunities that provides for the users to define how they want to run their applications, not to have the cloud tell them how their applications have to be run, and now you've got to rewrite all your applications to fit our model, you know. So. Well, certainly pure-played public cloud Amazon does well. Again, a lot of holes to fill. In OpenStack, within the community, Jonathan was on stage, just wanted to talk about the Quadrant, Niche, Unicorns, I feel like the top left was, but then the Winners, and that's adoption, technology maturity. Yeah. Virtualization, no problem. Core, compute, get that. Sure. What's next? Because stuff's moving fast up. What's maturing the fastest? Can you handicap some of the things? So I'm still extremely bullish on bare metal, which is, maybe shows my, it may not look like I'm an old gray beard, but I'm an old Unix gray beard. So I'm still extremely excited about the fact that we're doing more and more in the bare metal space and enabling that in the same control plane and connected into the same networking infrastructure. Those are kind of big deals. Also, I can't possibly get any credit for saying, oh my gosh, I think containers are exciting, because, you know, I mean, oh wow, big news. Yeah. Woo, I'm so amazing. But what I think is actually interesting there from an OpenStack perspective, isn't let's go reinvent what's happening with containers. Guys are doing a great job, but if I can hook that in again into the networking fabric, if I can make those things run inside, but it turns out some of your applications might want to run in bare metal, some of them might run in VMs, some of them are great for things like Kubernetes and Docker and Cloud Foundry and things like that. Not all applications are made the same and most applications can't work in all three. You got to pick where they're going to go. And so I'm really excited about getting the integration between those things. So that if you go full on Kubernetes, but you've also got this, you know, Oracle database, you've got to run over on a piece of bare metal iron in the corner, why not? Why wouldn't you do that? And why wouldn't you want that to work seamlessly? So I got to ask you the question. I'm asking every guest. What is the deal with Hybrid Cloud? Does it exist? You get private in public. You get private in public. I mean, is that like a product? Is that like a category? Help us create this conversation there, because Hybrid Cloud seems to be the top conversation. It's an outcome. It's an environment. But is it actually a category? Is it a product? Is it a mindset, paradigm? I think it's a mindset and I think it's an opportunity. So I run with another team of people and actually let's give them more of the credit because I'm, you know, I'm talking with you fine folks while they're off actually running something. And you're working the job, so. Yeah, so we run all of the developer infrastructure for the OpenStack project on top of OpenStack. And when we say that, what we actually mean is that we're running it on top of right now two public OpenStack clouds and one private OpenStack cloud. And then in addition to that, we have another private OpenStack cloud and another public OpenStack cloud that are on deck to get added into that. We run the things that we're doing across those as if it's all one set of infrastructure. So if you sort of think about it about people in their enterprise maybe have three, four data centers. You've got some different locations. Some of those may have different kit in them, right? So you may have the one where you do your high throughput things. You may have different characteristics. And for us we get to choose what things we want to run where or there's some things that can run anywhere and our scheduler can just place those workloads anywhere we want to place them. So. So the key is moving workloads. But it's moving workloads. It is, but I want to be clear what I mean when I say moving loads. There's other people who mean different things. I don't mean I'm running an app here and all of a sudden it's in Texas in Rackspace's data center and I click a button and all of a sudden it's in Paris in OVH's data center. That's not what I mean by moving workloads. I mean that to me creating compute resources in any of those locations is all the same. I can create one here, I can create one there. The private cloud that we're spinning up inside of the OpenSec infrastructure project is a cloud that is explicitly tuned for our workload. Now it has the same UI, it has the same, it has all the same API calls as our other clouds but for it we've gone and said we have a particular workload that we know very well because it's been instrumented in our other ones and we're going to just run the services we need in this private cloud. We're not going to have any other tenants in it. So we're not fighting noisy neighbor. We get all of, so we get a really good environment but we manage all of our resources, all of our IT resources in the same way and we basically got a file that says this server goes in this cloud, this server goes in this cloud and these servers over here go across these clouds and that's really great. So if I understand you correctly and correct me wrong. So the hierarchy is private cloud and public cloud are subordinate to hybrid cloud because hybrid cloud is ultimately everything. Hybrid cloud is everything. You want the freedom as a person running an application, as a person running an IT organization to understand your workload, to understand your application, to understand its needs and to be able to utilize the resources that you've got in the way that makes the most sense for those things. Again, this is back to the other thing. I don't want to say I want to rewrite all of my things so they make sense on a public cloud. Some apps don't make any sense that way and you shouldn't need to rewrite your things. You should have the infrastructure should give you the freedom to run the apps you need to run. And we're asking this question because it's a conversation we want to have because the definition has been kicked around and it's been mangled all over the place and more importantly, it's like distributed computing. You can't, that's the concept. The outcome of that is deployed infrastructure, stuff was running, stuff's being managed. I went to the store the other day and I bought some distributed computing. It was fantastic. I'm the leader in distributed computing. But what, is that a category? Who's the vendor? So again, separate vendors and products and talk about environment. That seems to be what hybrids. That's what it is to me. And one of the promises that I see in OpenStack is that in addition to that mindset, it should be possible to mix and match vendors. Like this is one of the great things here. There's absolutely no reason. I mean, in fact, the public clouds that we're currently using on deck, there's rack spaces and HPs, which I don't know if you know, but those are two different companies. We have a team at Red Hat running one of the private clouds that we use. That's a different company. We have OVH spending up resources and we're going to be getting resources from IBM. So like, those are all, and we can still spread our apps across that. That's fantastic. So why can't Amazon just say, we have hybrid cloud and put Amazon inside the company? They can say anything they want to. And I challenge you to tell Jeff Bezos, no, on that subject. Well, the CIA is a private cloud in Amazon, so it's kind of like... But that is a, so two things. A, that is a private cloud in Amazon. I heard that the effort and financials on that didn't work out quite as well as the Amazon folks might have wanted for that effort. And of course, I don't actually know anything because it turns out I am neither in the CIA nor Amazon. But you're hearing rumblings. I'm hearing rumblings, and I didn't, as I understand it, that isn't really a, if you think about what their core competency is, right? Amazon makes a profit on the sale of a 25 cent, you know, used novel, right? They can turn profit on that. Doing lots of custom bespoke work for a large vendor, that's not their core competency. That's not what they do. That's not where their margins go. That's not where any of those things go. So they can chase that, but why? Why would they chase that? What possible good does it do them to do that? I mean, their customers are asking them. Maybe their customers are asking them. So we'll see. But the thing is, if their customers are asking them, and you want to talk about companies that can deliver large scale custom bespoke solutions for large enterprise customers, you might want to go to a vendor who's actually good at that. Monty, it's great to hear from you. Great to get the data and the insight, certainly as a technologist, great insight. And the competitive strategy and the landscape between the opportunities is, you know, like moving the goalposts in the game. So we'll see. It's fun to cover. I got to tell you, it's been great. Excellent. As long as we can provide you fun, this is basically what I'm in for. People are tooling up, people are changing. So this real work of being done, wealth is being created, customers deploying new solutions. So it's very relevant. So appreciate the insight. Monty Taylor now with IBM, CUBE alum, friend of ours, great to have you on theCUBE. We'll be back with more live in Silicon Valley after this short break.