 Welcome back everyone, this is Silicon Angles theCUBE, our flagship program. We go out to the events, extract a similar from the noise. I'm John Furrier, the founder of Silicon Angles. Join my co-host, do Miniman, and also at Wikibon.org, and our next guest is Craig Muzilla, Vice President and General Manager of Middleware at Red Hat. Craig, welcome to theCUBE. It's great to be here. Great to have you. We had great conversations this week with all the executives, and the top conversation seems to be, well, obviously, pretty hot, popular product. Open stack, an open shift, really, really interest, a lot of general interest around that. So, is that surprising to you? And give us the update. Well, I mean, obviously the industry's going through a major transformation, and it starts with sort of basic cloud services. So, open stack as an infrastructure, as a software capability, it sort of starts there, especially when you talk about Red Hat, because we start from the infrastructure on up. But what I think is also interesting is, and our strategy has never been to rest on our laurels on the infrastructure side, but to start to move up the stack. And so, it gets into open shift, it gets into middleware, and how you build applications. I got to share a quote with you from yesterday. Yesterday, I just can't even keep track of all the great sound bites from the CEO and Cisco's on. But one quote that I liked was, you know, at Red Hat, we like to light up the hardware. And I'm like, well, now we're lighting up the clouds. That's obviously the next step. So, that's about reliability for the enterprise, and lighting up the commodity hardware business. And you guys really, and Linux created that decoupling. But now going into the cloud, software becomes the key thing, where DevOps becomes the key enabler now in the cloud. So, how are you guys going to light up the cloud? It's a great question. We run these customer advisory boards, and we have a customer advisory board called the Strategic Advisory Board. We bring in some of our key customers from around the world, and we had one recently. And we sort of asked this question, what's important? You know, how do you think about the cloud, and what's important? And they said this quote, and we've used the quote, actually Paul Cormier used it in his keynote, the applications are king. So, there's a lot of hype and buzz around the infrastructure. But at the end of the day, if you're a CIO, you're thinking about, how do I create and deploy an application more efficiently and faster than anybody else? And so, yeah, it trickles down to the infrastructure in terms of scale out, but it actually begins with the app builder, and the application development as well. You know, I thought you had a lot of the conversation around app acceleration, app delivery. It's a really interesting one. Obviously the containers is a great message. But you're in the middleware area, which has always been a playground of innovation, but also co-optition and competition. Yes. You could lay out the vision around the middleware market in terms of competitors, frenemies, partners. It's a pretty dynamic time in cloud. A lot of people have, a lot of use cases, not one. It's all, so lay that out for us. So, we started getting into middleware around 2005, 2006, but we got real serious in 2006 when we acquired JBoss. Now, JBoss had built a name for themselves in producing a J2E application server. So, the reason we did it as a Linux company, as an operating system company, was because we knew it was important to have a relationship with the developers and pull in workloads onto our operating system and our infrastructure. But after that acquisition, it was clear to us to be strategic in middleware. You needed more than just the app server container. You needed a full breadth of middleware capabilities. So, we've been building out either some organic stuff in the community or through acquisitions that we've done. So, we now have 12 different products in the portfolio. I got to ask you a question about DevOps because a theme of the new generation is DevOps. And sometimes Red Hat doesn't take enough credit for really being a DevOps enabler. I mean, if you do Linux, that's ops. JBoss, that's dev. When you're together, you got DevOps. But DevOps has come from more of an application lifecycle management kind of mindset, right? Going back to the 90s, right? So, ITIL, information technology libraries, service-oriented architectures, that was the beginning. You guys were a part of that on the Linux side. You bring JBoss into it. So, in a way, DevOps has been a part of your heritage from the beginning, right? Yeah. That's a great way of looking at it. I think if you say, look, our core was operations and infrastructure, but now with all this middleware, we're bringing in the dev component. So, what better vendor, especially from an open source perspective, is that to bring together and have a DevOps solution, middleware plus cloud, middleware plus infrastructure equals DevOps. And most of those DevOps guys are command-line guys. So, okay, so now let's go back to the JBoss, since then and now, what has been the big change that's enabling all this conversation? Because now, DevOps, you're talking about Node.js, you're talking about real-time applications, you're talking about real, different-looking stacks and frameworks, that's not your blocking and tackling lamb stacks. Right, right. So, what has changed and what have you guys learned that give you a unique position today there? Well, we started embracing the concept of allowing people to develop in a variety, we call it open choice. So, it wasn't just about programming in J2EE, but it was supporting other frameworks such as Spring, but also supporting a variety of other languages, whether that's Groovy or Ruby on Rails. Now, when you bring in OpenShift into the equation, we stand up Node.js, you can program it in JavaScript. So, it's really providing developer choice in how you go about developing your apps. And that's a big change, I think, over the last seven years or so. So, Craig, you talked about there are some services that are developed by the community and some things that Red Hat acquired. How do you balance that waiting for it to just happen organically versus moving a little faster or finding the right fit that might make sense to bring into Red Hat through an acquisition? It's a great question. I think we look at various categories where there might not be a lot of innovation happening outside, maybe there's no innovation happening in Apache for whatever reason. We will start a project. So, for instance, we started a project called InfiniSpan, which provides a distributed data caching capability, a data grid. And so, there wasn't a rich alternative out there in other communities, so we started that community and now we have a product in that area as well. In other areas, we made an acquisition going on almost two years now of a company called ViewSource. They're a leader in integration technology, so James Strake and Rob Davies were some of the principals of that company. They created the very, very popular Apache projects, ESB, Service Mix, Camel, ActiveMQ, Craft to manage all fabric. So we saw a really vibrant community that was moving way very quickly, advancing beyond the standard technology providers, and we said, look, we need to bring those guys in, we need to be part of that community and build around that and take advantage of the momentum in that community. So, with your open source methodology, is that limiting at all because in some ways, there's lots of software companies that could be out there and might have some cool innovative feature, but you really require some kind of community to help build it, so is that a limiting factor for some of the services that you might want? First and foremost, we're an open source company, so everything we do, we have no sort of tricky policies where it's a bait and switch or we don't do open core, everything is open source. So, when we look, we have to make sure that we can do something in open source, but that doesn't mean to say that every acquisition or everything we do starts with open source, we have made acquisitions in special middleware areas. For instance, we just introduced a business process management product. We acquired a company in Spain called Polymeda, they had a great web-based BPM product, and when we bring them in, it was small enough that we could create a community around it, take it open source. So, yes, we look for open source DNA, or what we might, even if it's proprietary to begin with, then we figure out, okay, how are we going to create a community out of it? So we don't necessarily, the prerequisite isn't necessarily, oh my God, it has to have a big community to for us to do anything with it, but we have to be confident that we can create a community around it. Great, you own both JBoss and the OpenShift solutions, can you talk to us about how those two fit together and are they tied together? Does one require the other, or is there some flexibility there? So, OpenShift is our platform as a service, and we have an online version called OpenShift Online, and we deliver it as a product that you can stand up a private pass called OpenShift Enterprise. When you look at the state of paths today, it's still pretty simple. You have some basic capabilities, it might be MoJS, it might be PHP, Java containers that allow you to create a basic Java app, but they're, I like to call them baby apps, they're small web apps, right? Most enterprises that are creating something more complex need a lot more tools, they need an ESB, they may need business process management, they may need caching or data integration. So, the idea of a reference architecture for middleware brings in all of these various pieces, so there's a couple of analysts that have been using the term XPass, and so you start to see specialized paths as you see APass for building basic applications, but now you're seeing the term iPass for integration, ALM paths for application lifecycle management, you're seeing BPM paths. The idea is that you need to bring to create a true enterprise application and have that type of adoption in paths, you need to bring all these capabilities in. So, in September of last year, we introduced a concept called XPass and our vision beyond that, called JBoss XPass for OpenShift. In the last, at this event, we just announced that we took live some of these first services, now we took the Fuse technology, we have an iPass service, we have a BPM path service, we have a data virtualization or data integration service, and we have a mobile service. So, I think we're in a very unique position with the rich portfolio that we have to bring in all of these various pieces of middleware to create true enterprise applications, to move beyond the baby apps, as I like to say. So, big data is always a conversation we have and we've asked everyone this question, it's been kind of like, I won't say dogs, there's really no right answer, we're trying to tease it out as that, obviously storage, obviously a piece of it, we talked about storage. Where does the data layer fit in there? Because it really hasn't piqued its head out yet to kind of figure out where that is, people have different approaches. Certainly, big data with Hortonworks, you guys have a relationship with, the red hat at Dube, if you will, but if you guys have the red hat of OpenStack, which is happening, we see that more and more the reality. Where does the data fit in? Because you're seeing the bell, whether it's Facebook, LinkedIn, Twitter, guys who use data as a competitive advantage, it's almost a developer resource. There's an ecosystem building around that, so it doesn't seem to fit, like it's not a clean subsystem in the overall operating system. What's your view on that? Well, data's always been a separate service, right? So whether it's a relational database, whether it's some of the no SQL solutions today like Mongo, whether it's big data concepts for doing analytics like it do. We are adding the data services to OpenShift, so we have relational services today. We have no SQL services available based on Mongo, and so we're continuing to build that out. We have some other unique technology that actually helps along. We have this product in the JBoss portfolio called data virtualization. And it allows you to, we like, there's some work that we're doing with Hortonworks right now. The campaign's called Making Big Data Real. So you've got all this data in Hadoop. You went right to where I was going to ask the next question. You've got all this data in Hadoop, but you also have three or four databases over here, and so how are you going to bring together this unstructured data that you've collected in massive amounts in Hadoop and bring it together with some structured data in a relational database? What data vert allows you to do is it allows you to set up an abstraction layer, it allows you to query against a Hadoop store, it allows you to query against relational databases, bring it together and deliver it in any format you need. Yeah, we like that trend. I think that's a really big trend. In fact, we had Padma say on theCUBE yesterday, she thinks that data virtualization is a really big deal and a next big thing because you can apply it only to the applications. So whatever application you're writing, whether you're, whatever you are on the stack, this is a real benefit. Right. Do you agree with her and how would you explain data virtualization to the average journalist or the average industry guru who's not on the trenches? So data virtualization is about, okay, I've got three or four different buckets of data and the formats are all different but my application needs to see data from all five buckets or four buckets. Data virtualization allows you to put this middleware layer in between and you can ask all five buckets to deliver data to you and you get it in the format you need. In real time. In real time. You can see really pinging the data, not sending it out into the data warehouse sometimes. That's generally how it's compared. How do you deal with that type of problem without data virtualization? Well, you copy the data, you go through an ETL process, you dump it into a warehouse, you dump it into another data store and everything's out of sync. Anything that changes breaks the application. There's a quality issue, there's lag, all that stuff, so that solves that. And how much is that hitting the main street right now? Is we tipping the toe and the water at this point or how would you put that in terms of where it's at on the market? The data virtualization types of concepts and technology has been in the marketplace six, seven, eight years now. I think because of big data and because of cloud, there's going to be an acceleration. And the reason being, take cloud for example, I've got a private platform as a service or maybe I'm doing something in public cloud and now I need data that's behind on a physical system, behind the firewall. How am I going to get access to that data? So DataVirt actually helps you bring in some of that data. And also there's just so many really wild use cases, just data governance, policy, all that stuff. Okay, so that's exciting, we're totally stoked on that. So I got to ask you a question about on a personal level, what are you most excited about right now within Red Hat, what technologies, activities, things about Red Hat that really get you excited and what outside of Red Hat are you excited about? That you're watching very closely. Oh, great question, where do I start? I mean, I just think that Red Hat, we're in an enviable position in the marketplace that we're coming from both perspectives. We're coming from an infrastructure perspective and having the strength of RHEL and now doing OpenStack and we're coming from a perspective of having a rich portfolio and capabilities to do application development. So your notion of DevOps, bringing them together, there's maybe one other vendor in the marketplace that has that breath of the asset of operating system and infrastructure to application development. It's not like you're making it up either, it's not like you're saying, we're doing marketing, we should do some Linux and DevOps. Exactly, we have it, we've been doing it. It's like, hey, marketing guys, why don't we put that together? Yeah, okay, so outside of Red Hat. Clearly the big data, well, everything is happening with cloud is obviously first and foremost in everybody's mind. I think the jury's out in terms of how this will actually play out, we have our own views on that of course, so that's very intriguing. I think concepts like big data and what eventually happens, is there a sort of an evolution to replacing sort of traditional data stores and relational databases with other ways of doing things? And I think that's something that we keep our eye on quite frequently. And your take on the whole container thing, that's all the rage right now, obviously Docker, good timing. Container's not a new concept in technology, but like right now, it seems to be a hot area. What's your take on that? Well, it's very, very important. I think it's also part of the evolution of the marketplace. Like everybody's been talking about virtual machines. Now we're talking about virtualizing the app, virtualizing the image itself and gaining the advantage of portability across all these environments. If you heard Brian Steven, our CTO's keynote today, talked about the density improvements that you get by starting with the container at that level rather than dealing with virtual machines. I think it's an evolution. We had virtual machines in the 2000s, now we're in the 2010s or 2000 teams and Docker is exciting that the technology and I get the middleware is going to be all, I think the battleground for all the innovation because it's going to be fun to watch. I got to ask you more on the general management side as an executive, you got to make money. You guys got a couple of company of stockholders, stock price to manage, but at the same time you have this massive communities that you're managing. How do you balance the transparency around, you got to make money with the upstream support around all these emerging areas from enterprise grade. Certain things you got to see with reliability to the wild west of apps men, kick ass, take names, kind of mentality on the app side. The DevOps culture as they come together. You got to make money from that. How do you deal with the upstream and pull that down and make money? So we've always, it's a tricky balance. There's always a balance of okay, we got to package something up and make a product and we want the innovation of the community. We've done some things recently, even in Red Hat with the JBoss communities, which is to create a little bit more separation between the upstream. So for instance, we changed the name of the very, very famous JBoss application server to Wildfly. And one of the reasons that we did that is we wanted to see more innovation and a faster pace of innovation upstream and we expect Wildfly to be moving at a really fast clip. We'll take those things downstream, we'll harden them and then we'll release the JBoss enterprise application platform, the commercial app server. So you have stability downstream, but upstream it's WildWild West, we encourage them and we're trying to do more of that to encourage that. My final question, I know we're getting a good segment here, so we'll get into the day, get a little pump of energy, shot of energy. Tell the folks, Craig, why is this show so important at this point in time? In this history, obviously 10 years you've been doing this event, Red Hat certainly has been around for a great brand, but why right now? What is so important right now that people should be mindful of it? Well, there's so much change in the industry right now, so an event like this is a big educational, it's an opportunity for people to learn, people to contribute new ideas. I've had so many conversations about XPAS and some of these other new concepts that are coming out. These concepts weren't even discussed two years ago, so an event like this is very important to learn, to hear what's going on, to influence the industry and I don't think I've seen this much change in the IT industry probably since the mid-90s, 95, 96, when the web applications were really hitting their stride, this is a great time to be in the industry. People are building out and bulking up, they're investing in growth, not just consolidation. It's like, hey, let's not consolidate those servers, I mean, I'm oversimplifying it, but that's where we live for a decade or so back before. This growth curve, now it's like, hey, get that web app out, and the iPad was a seminal moment to me, right? When CIOs saw the iPad, they say, I want that, wait a minute, why aren't the astronauts? And then he goes, oh boy, wait a minute, that's our infrastructure? What, so where's all our developers? Oh, we outsource them all, I mean, I'm oversimplified, but that's kind of been the state of the industry, but now the mandates are business drivers, top-line revenue growth. That's, has not been an IT conversation until recently, right? Well, it's interesting, I think many companies now are defined by their apps, by their software. Software was always sort of a supporting role, right? Yeah, you had an e-commerce site, but it wasn't, if you didn't have the e-commerce site, would you be out of business? Now today, the apps really define the business, so that's a big change, I think, in the use of technology by companies. And now that the consumers are connected, fully instrumented in the technology environment, internet of things, I mean, they've talked about big data tsunami, I mean, that's coming in fast, I mean, if they're not already full and bloated with data, I mean, that's why I like this data virtualization, I think that is really something that I learned in this show, that I was watching, but to me it's now clear that some sort of data virtualization first mindset is going to be upon us. Yeah, just deliver the data, I don't care, you know, it's coming from many different places, just get it to me. Craig, thanks for coming on theCUBE. Craig Mozilla, Vice President, General Manager of the Middleware at Red Hat, here inside theCUBE, breaking it down for us. We'll be right back after this short break. Great.