 Okay, we're back here inside the cube. This is SiliconANGLE's exclusive coverage of OpenStack. This is the cube where we go out to the events, then extract the signal from the noise, and we've got one of our favorite companies that we like these days is Hortonworks, mainly because, one, they're in Palo Alto. Two, they are in the Hadoop space, which we've been covering since Hadoop Summit, second year of Hadoop Summit when SiliconANGLE started in Cloudera's office. And you've got Cloudera and Hortonworks, really the two main guns that we've got. Cloudera and Hortonworks, really the two main guns in the open source. Hadoop, in terms of pure open source, but you guys are a little bit more pure than Cloudera. Jim Walker, who is the director of product marketing here at Hortonworks. Guys, I'm really impressed, you have an announcement today, so let's get right to the news. Hortonworks is Hadoop. It's open source. The entire team has legacy in open source going back to the days of... All the way back to Jboss. Jboss and Spring Source and all that stuff. So you guys know a little bit about open source. And here, the success of OpenStack has just been amazing that they really shielded up the hype, kept the kernel of core dudes together with code, and really is a shining example of success of open source. And the continuing drumbeat of our editorial, which is Scale Out Open Source wins in this new market of modern infrastructure. So give us the news of Hortonworks here at OpenStack. Yeah, first of all, I mean, this is amazing. I guess I'm kind of new to the whole OpenStack world. I can't believe there's another space that's as wildly as exciting as Hadoop right now. It's kind of like this whole other world going on, right? So this is great. So what we're doing here is, we feel that Hadoop is just a really powerful source for adoption of OpenStack, right? We feel that it's kind of almost the perfect app to help aid adoption of OpenStack. I think if we look at organizations the way they're using servers, the way that adoption is happening, it's really following the workloads. And Hadoop is really kind of that net new workload that we're finding organizations adopting throughout. And so if you think about that and start to say, well, how am I going to implement infrastructure for that, Hadoop makes a whole lot of sense. Now Hadoop needs the cloud from an agility point of view. If I could spin up clusters, spin clusters down, be more agile with my Hadoop environments, it's hugely important for Hadoop. So it's a really good fit. So you guys announced something with Red Hat and Morantis? Yeah. So the announcement really is to help aid that, right? And so working with Morantis started a project called Project Savannah. It was a couple of weeks ago they announced that, right? So we've partnered with Red Hat and Morantis to really, I guess the words are, contribute significantly to really build that out. Savannah's really all about simplifying management and deployment of Hadoop on OpenStack. Hadoop's not simple in itself to deploy. It's an operations point of view. It's not fully simple. And then you have the complexity of OpenStack as well. So we really want to make it almost push button. So taking a Hadoop cluster, say I want to run an H-based cluster for a test, being able to provision that very quickly via templates or blueprints of what a cluster is, really what we're kind of all after. Well, let me just parse through that. I'll just say it. Hadoop is hard to work with. H-based in particular is a lot harder because you've got to have the expertise and you've got to host it, where you've got DynamoDB on Amazon kicking butt right now. And also they've got Elastic MapReduce on Amazon. So you're seeing that people want simplicity and they want choice, which is kind of like, I'm not saying OpenStack. I mean Hadoop is hard, hard to use. For the average developer, it's got to get simpler. There's a lot of configurations, John. I'm not going to sugarcoat it. There's a lot of configurations when you deploy a cluster and to optimize that for the right performance. There's a lot of things that you're doing. So how do I make that push button? Because the infrastructure guy doesn't want to have to worry about configuring all that. And so really that's what production is all about. And it's key. It's key for adoption. I think that is the key thing. And I tweeted earlier. I said Hadoop presents at OpenStack. Oh, that's the wrong tweet. Hadoop is a non-legacy app. I want you to comment on this. Hadoop is a non-legacy app well suited for scale out OpenStack architecture. Still lots of work to do, but opportunity for developers. Juan, do you agree with that? Or you must, because I see the open source piece. But what is the opportunity for developers? For the developers, I mean, just the combination of the two. I mean, it's the net new workloads that we're having in organizations, John. I mean, if it's, oh, excuse me. That thing's kind of coming off there. You know, if we think about net new analytics, the whole kind of move towards big data. Mobile. They're kind of moving along in the same path. So I mean, I think it is the perfect app for OpenStack. I mean, it's, like I said, it's driving those workloads. And I think that's where the opportunity is. But you have to simplify it, right? And so our goal is to really fuel and foster the adoption of Hadoop everywhere. It doesn't obviate anything. It's a plus one in architecture, right? What are you guys bringing to the table? Code? Are you guys bringing more resources? You're hiring? You've been to throw some people at the OpenStack summit? You mean to the OpenStack world? What do we bring? Yeah, code. I mean, you know, I mean, you know, this doesn't get done without a lot of work. I didn't mean that. We know you guys are putting a big dent in the universe on Hadoop. You guys are doing amazing work in the Hadoop space. And you guys are pioneering it. And actually getting a lot of, you know, I think you might even be in the lead, as far as, you know, Dave and I, Dave Vellante and I debate that all the time. We think you have a lead over, over Cloudera right now. And certainly them not being here is kind of a diss on Cloudera and kudos to you. But specifically to OpenStack, because that's where the enterprise is going to start to get around that. It's a great go to market for you guys to get in the enterprise through this community. And so, you know, what are you guys going to bring to the OpenStack? Just code. I mean, code and expertise. You know, I mean, like I said, you know, there's the complexities of Hadoop in itself. There's complexities of OpenStack. You know, it's the beauty of the open source world, right? The fastest path to innovation is open source. You know why? Because you can get all the right people in the room to code the right thing. And that's just, you know. I'll follow up with that, Jeff, because we had a conversation a week or so ago and we were talking really about the innovation. But now this is kind of the second order of open source, where you've got kind of an integration of Hadoop and what it represents in terms of storage and the big data. And then you've got this cloud infrastructure. And so kind of being really intimate with one and jumping in with both feet on the second. How do you see, you know, kind of a combination? And there's another one we heard down with the networking guys down in San Francisco today. So now that there are more kind of open source foundation clusters. I don't know if that's the right way to describe it and how they interact and how is that impacting how you guys view the world and where you spend your time and resources. Yeah, it goes back to kind of where John was going in terms of the intersection of how these things come together. And if, you know, if you look at the three companies in our release this morning, it's Red Hat, Morantis and us, right? So you have Red Hat, who is the leading contributor to OpenStack within the Grizzly release, right? So more lines of code than anybody. You have Morantis, who is definitely one of the leading system integrators within, you know, the OpenStack world, right? So now you have the guys who built it. You have the guys who implement it. And now you're bringing together the, you know, the leading contributor to Apache Hadoop and all the related projects. I mean, anybody could do the math around code counts and committers. And I'm not going to get into that kind of debate, but I mean, you have the three different groups coming together to actually promote now simplifying Hadoop on OpenStack. It's bringing the right people together. We were talking about this last week. It's just so much, it just makes sense. You get all the right people in the room. It just makes sense. You're going to be able to get there a lot faster than a proprietary firm. And you put the right services and support around that. It's just a really good business model. Did you hear the Red Hat thing earlier? The Brian Stevens say, you know, open source is a development model, not a business model. It is. It's a development model. And if you think about that. I actually think, I think, I would say both. I mean, they're not mutually exclusive. I mean, look at Rackspace. Rackspace's halo effect. They moved the ball down the field in a matter of three years with OpenStack. I was having customers with Jim Currie. He'll be on tomorrow. And Lou Mormon, the president at the time of that division when they had cloud sites. They could have owned this thing on their own. They could have completely bow guarded this community and tried to kind of throw a little bit of a red herring community out there and take it in house. Because they needed to build. What OpenStack did to me is an example of the business model open source. They said, you know what? We're going to be completely transparent. We're going to put and contribute resources. And the tide will rise all the boats, being the biggest boat in the harbor. And that worked. Look at them. And they're just significantly more scalable. They're leveraging all the work of the people in the community. That's a home run. And John, that last word is the key, right? So there's open source. Anybody can push code out in the GitHub. That's open source, okay? I mean, building a community around that and incenting that community to actually contribute and take part in it. And for the common good of everybody, that's not easy to do. And it takes... They walk the talk. They walk the talk. They walk the talk. And they brought code to the table. And that's the ethos here in the OpenStack community. Bring some code to the table. And you're in. It's the cover charge. That's the question, though. Everyone seems... It's been a consistent theme. Everyone we've talked to over the last couple of days is, wow, look at the size of this community. Wow, look at how fast this has grown. Wow, look at... By luck or happenstance, the governance that's been put in place to enable this just really rapid adoption and growth. So from your point of view and with more experience in another community, what is the magic pixie dust? What is it that's really given this thing the juice to grow and be as successful as it's been? Passion, dude. Look around. Look around you. Everybody here is super passionate about this stuff. Every session I've been to, I don't know if it's because the rooms aren't big enough, but there's people waiting outside to go and watch this stuff. Developers are passionate. I mean, I'm an ex-developer. I was a developer for eight. Yeah. We go weird. It's like... When you're into it, you're into it. Yeah. Jim, our expression with theCUBE is we go to the events to extract a signal from the noise. Well, there's too much signal here. We need another cube. There's not a lot of noise. And David Floyd and I were talking about that. Really? There's not a lot of vendor hype here. You get the marketing, some PR going on. You have some releases going out. The booths are modest. It's really all meat and potatoes. It's all code, user-based customers. And the guy from OpenStack put up that slide earlier. He's like, here's the platform. It's great software. Innovative ecosystem and successful users. And that's the holy trinity of open source. And they're demonstrating it. It's a nuance that really can't be lost and it's super important, John. And if you're going to build a company around open source and Hortonworks, we are all about that. I mean, it's really about incenting that community. It's about being friendly to the developer. I mean, we think every day about not being overly marketed to people because it's... The community is absolutely important to us. We are a huge part of that community. You guys are very transparent. I was very critical of Hortonworks when you guys launched out of Yahoo because I was at the time, Cloudera was like the shining star. And it looked like kind of a head fake. What's going on Yahoo? And within six months, it was very clear that you guys were putting all the cards on the table, a great group of guys. And you stayed true. And Kudos to even MapR, right? So MapR, again, another great company in the open source base. They never really head faked anybody. They're like, hey, you know what? We're enterprise. We're going to check all the boxes. And we're just not them. We're enterprise. We'll use Hadoop. And they stayed on their course. So I think that is really the key. But I want to ask you about OpenStack. What do you think from a biz dev standpoint, product market standpoint, the success of the OpenStack will be in the big data world. Because at the top of the stack, you have analytics, mobile, and a lot of big data applications. What's your take on that part of the stack? You mean from an OpenStack? Yeah. And what needs to be checked off and what needs to be worked on the white spaces? Yeah. I mean, from a net new analytics to a Hadoop to OpenStack kind of configuration, there's a lot of talk about, you know, how does HDFS work with Swift? And, you know, am I storing in TrivaKit? There's a lot of issues that have to be kind of solved from a nuts and bolts around the stores side of things. Not our biggest concern, I think initially. Initially, we're more focused on kind of making it easy to deploy. Because it's two groups of people. You know, you have analytics who are business users and you have ops and then you have this infrastructure guys. And they don't talk. They just kind of like expect each other to work. They really make it simple for the infrastructure guy. And that's, you know, bringing Hadoop expertise into OpenStack and making it work. And that's what Savannah's all about. I mean, I think you guys have the Marantis guys on next, right? Yes, he's right there. The CEOs are coming on right next. And we're going to talk about the service. And obviously, to manage services side, it's huge. I got to ask you about HBase. Okay? We're a big fan of HBase. We actually use HBase and some of our code that we use for our collection of listening tools. But one thing about HBase is you know, like a tailored suit. Once you do it, it fits great. But it's got to get easier. Any update on you from Hortonworks on HBase, easier because when you get to a cloud environment, you've got to be versatile. Can you comment on that? Yeah. So, I mean, you know, the HBase community is absolutely moving very fast as well. So there's a lot of things going on. The HBase community make it simple. And that's from, you know, from the Apache side of the world. I think what we're doing with Savannah just fits very well into that. It goes back to this whole, you know, if you're going to use a piece of Hadoop, you've got to go through all these configurations. How do I make that simple? How do I create, you know, excuse me? How do I create blueprints so I can just knock something out very quickly? And that's the part of the management and monitoring of a Hadoop cluster that's actually super, super important. Because once you optimize what your HBase environment looks like, how do I actually then, you know, take that and replicate that or use that over and over again? And that's blueprints and templates become extremely important for that. And so, you know, integrate that within horizon so that within horizon you're kicking off, you know, the deployment of a cluster. Let's make it simple. So how's Rob Bearden doing these days? Because I want to ask you a question about a question and not ask him if he was here. The business model of Hortonworks. And you guys obviously are still doing your thing. So I want you to give the update on what's new with Hortonworks as a company. You know, not the open stack stuff, because we just covered that. And then the red hat for Hadoop has always been kicked around as Cloudera had that label on it. And then there's been quite a debate. We had Pat Gelsinger on the Cube two years ago saying there'll never be a red hat for Hadoop. Madisay and I got on a little thing on Twitter about, you know, can there be multiple versions? I mean, you know, is it a debate? Is it, what's happening? Help us dissect that. You know, I mean, nothing's changed with Hortonworks. I mean, we have stayed true to our mission from day one. I mean, we are all about fueling and fostering the adoption of Hadoop in the larger enterprise. We don't want to obviate up the stack. We know where we fit. You know, Hadoop is part of your overall data architecture of the future. And we know where it fits. And for us, it's making sure that everybody in the world uses Hadoop. And so we are marching on that path. The only difference between, you know, probably when I talk to you or anybody's talked to you five months ago as opposed to today is the rate at which customers are coming to us and paying customers. Not just people using HGP. You know, it's across every vertical you can think of, John. It's financial services firms. Of course, a lot of retail guys, a lot of telecom guys. So I mean, it's just, you know, it's dozens and dozens of customers now. And your business model is still service. And it's still services and support. So, you know, we aren't selling software. We aren't building proprietary extensions on top of Hadoop. We are pure producers of Hadoop. There's, you know, in the open source world, you have packages and producers. You have guys who use the open source core and build on top of that. That's the beauty of Apache. We love that, right? And then there's the pure people who are actually producing. We produce. And so, you know, our game is to make sure that really Hadoop becomes that enterprise viable data platform. And in order to be enterprise viable, it's got to be stable. It's got to be reliable. The enterprise has to trust it. You can't hire all these people to manage a thing. It's got to be simple. The part of the stuff we're talking about is simplicity. And trust. And that's what we're all about. So, you guys are getting some good buzz here. I tweeted that earlier. But Cloudera is not here. You're a competitor. Can you talk about how you guys are competing against Cloudera now? Is it still the same? Can you just give us an update on, you know, the Horton versus Cloudera? And I thought I want to get into a little war there. But, you know, it's important. You know, you aren't going to bait me into saying anything really bad. But I mean, you know, I mean, we love what the Cloudera guys are doing. I mean, they're out there. They're doing the drum beat around Hadoop as well. You know, they've got a lot of guys, really smart guys on the street doing some things. They're doing some great things with a lot of really cool projects out there. Do I like competing against somebody? I love competing. You know, it just, it makes the whole market better. And I think they're a formidable competitor. And they're doing great things. So, you know, we, I like to win. And we win a lot, you know, against them. And I'm sure they've won against us, but... That's a non-answer. Come on. Give us an answer. What do you want from me? I mean, you know... But they're open source. It's a knife fight. We don't... They're not here to defend themselves. I have to defend them for you. They're a lot of guys I like. So, there's a lot of guys on that team. I feel like, you know, the funny thing is about an open source community. They've got a lot of committers at Cloudera, right? And we've got a lot of committers at Hortonworks. You know, when they all get together, it's a Hadoop summit, you know, coming up and later in June. It's funny. They're all friends. They all talk about their wives and their kids. And so, you know... We love Cloudera too. Director of Product Marketing here inside the Cube. Hortonworks. Again, you've got Cloudera and Hortonworks. The two main horses in the open source Hadoop. Pure world. Thanks for coming on the Cube. Great to be here. And great move. I think Hortonworks is a great enterprise. In addition to your portfolio. Huge. And I think making this developer community and fusing those together gives you a leg up on Cloudera. I think it's a major fall for Cloudera not to be here at this event. I think they're missing something here. And so, Cloudera, you know, pay attention to OpenStack. We are here at the OpenStack Summit. I'm John Furrier with SiliconANGLE. We'll be right back with our next guest after this short break.