 Live from the San Jose Convention Center, extracting the signal from the noise, it's theCUBE, covering Hadoop Summit 2015, brought to you by headline sponsor Hortonworks, and by EMC, Pivotal, IBM, Pentaho, Teradata, Syncsort, and by Atunity. Now your hosts, John Furrier and George Gilbert. Okay, welcome back everyone. We're live in Silicon Valley, a breaking down Hadoop Summit behind us. This is day three of our wall-to-wall covers of Silicon Angles theCUBE, our flagship program, where we go out to the events and extract the signal from the noise. I'm John Furrier, my co-host this week. It's been George Gilbert, our new big data analyst at Wikibon.com for Silicon Angle Media. And it's been great. Our final guest is Sean Anderson, with Data Services at Rackspace. Good to see you again. Welcome back. Always a pleasure, John. Thank you. So, we're the caboose of the big train of Hadoop. The headline. Yeah, the headline, I'm loving it. Stamp on this show. The cleanup hitter. Yeah, exactly, exactly. We're going to just put a bow on this thing. First, what's going on at the show? What's your vibe? What are you seeing here? You know what's going on at Open Source. We talked at Percona Live, OpenStack. Buzz is here, but this kind of like, this low of like, you can feel the cloud coming. Cloud's coming fast. More compute, more DevOps, more automation. What's your thoughts on what's happening here at the show? Yeah, absolutely. And it's great to see the evolution. I mean, I think this is our second Hadoop Summit and of course Strata and just seeing the evolution is really amazing for us. But, you know, in terms of the cloud conversation, it's much more concrete now. We're seeing real use cases there. We're seeing real creative thinking around how to do data governance and isolation in the cloud. And it's really those forward thinking companies that are just, you know, blasting through the barriers of, you know, what people are telling them they can do and how they can architect these, you know, complex data structures in the cloud. It's really kind of fueling that. It's exciting for me. I mean, personally, we've been covering all the Hadoop conferences from day one and being president and creation of this industry. There's the, you know, drinking the Kool-Aid, you see the vision and like, yeah, let's all, you know, pump this hard, get it to going. And we hope it all works out. Then it's like, wow, this is working out. And then it's like, hell, this is worked out. Now we have a whole nother level of work to do. And, you know, the cloud thing is interesting is the timing is kind of perfect. So with that software becomes the critical piece. So how do you see the Hadoop ecosystem in open source moving quickly to fill that white space of analytics sistering up against these, you know, resource bases? No, I think, you know, fast moving is really the word there, right? You know, whether it's the speed of the versioning that we're seeing inside the Hadoop ecosystem or the kind of, you know, parlaying between distributions to kind of maintain relevance and direction with their customers. And I think that that's really driven by, you know, the demand, you know, it was interesting to see two years ago where, you know, one minute query time was like, you know, the monolithic end goal. And now, you know, we have customers requesting sub-second query times and, you know, not even micro batch but full streaming capabilities. And, you know, I keep thinking to myself when I'm having these conversations, this is a conversation I never would have had a couple of years ago. So as we increase the capabilities of the technologies, the demands on those platforms become increasingly more. So I think the ability to, you know, provide that feature parity to inhibit the roadmap to what their customers really want and what the power users really want is really kind of bringing on this V2 evolution of what people are trying to do with. It's not about, you know, the single use case, let's prove that out. It's like, I want to prove a use case while architecting my next use case and now starting to bring other open database platforms into form. Are you seeing, when you talk about, you know, going beyond the single use case, how does that get reflected in the infrastructure you're building? And by infrastructure, I don't mean, you know, just the hardware. I mean, like the polyglot persistence you were talking about. Absolutely. And so I definitely think it slows things down. I mean, especially with Hadoop, people struggle to get a reference architecture that speaks directly to their use case. And then when they start adding that in, say they go to more streaming or in-memory workloads, that drastically changes the underlying reference architecture. So for a company try to do it themselves, they really struggle to keep up with those demands. You know, one conversation that's going on and you mentioned polyglot persistence is, do we build the Hadoop application to solve all these problems? Or do we really start integrating, you know, an open source community of databases to fulfill those needs? And I think one involves a lot of heavy lifting on the development front, and the other one involves a lot of integration, a lot of hand-holding and cooperation between, you know, multiple database vendors. But I really like looking at some of the forward-thinking companies that really had no direction in doing a type of polyglot architecture and just went out there and did it. And they're building their own connectors, they're building their own tools inside of there. Sometimes they're being afforded to go out there and create companies based off the work they've done. And I think that's really amazing. I mean, they're going at lightning speed. How about rack space? What's going on with you guys? And honestly, the business is shifting for everybody. Sure. There's growth cycles, inflection points, as Rob Bearden says, how are you guys taking the advantage of it? What's going on rack space? We are just going lightning fast. So we've created a business unit inside of rack space completely and solely dedicated to databases. About a year ago, we rebranded as a managed cloud company. And that was based off a lot of verification from the industry, from analysts across the board that said there's infrastructure as a service and then there's managed services. And there's this new portion or new bifurcation of the market called managed cloud. And so a lot of what we think about is what does managed cloud mean to a database customer? And what we're finding out and getting great validation on is that open source databases are a great example of how managed cloud matters. And when I say that, it's very easy to start. Open source software free, download on a laptop, launch in the cloud, very easy to get started. But once I start to scale these and really put some high demands on that, that's where it becomes complicated. So these companies really grapple with how do I find somebody that's an expert in a two year old technology or a four year old technology? How do I build out my capabilities internally? If we can remove the kind of barrier into them, allow them to adopt these open source data place platforms in a speed that makes sense for them, then we're really kind of rising tide, lifts all boats and making sure. And there's also the enterprise grade stories not just for the high end Fortune 10s or Fortune 50s or Fortune 100s or for the Fortune 500 for that matter. It's for all your customers, you guys have brought up through the hosting, enterprise grades means reliability. So how do you guys talk to those customers that just say, hey, does it work? What's the pitch? That's an interesting question. So I think the easier adoption is the guys that are just building net new apps that have Hadoop as a back end or Mongo as they happen. They have a single use case, they understand what they want the technology to do. When we see larger organizations, they want to adopt some of those models. They have legacy systems that they have to incorporate into that. And so it's considerably slower for them to adopt it. They have concerns that these single use case companies just don't have. So a lot of what we do is take the learnings from those smaller companies and incorporate it to, how do I potentially port that into a large data warehouse environment? Or how do I transition from being a very legacy driven shop to doing some of the competitive things, especially around Hadoop, that are business differentiators? And how do I get that in there quickly? And how do I get value quickly out of it? Yeah, so we have one minute left, Sean. Appreciate you coming. I know you have a busy schedule coming in. Final word, what's next for you guys? What do you got in the marketplace? Got activities going on? Meetups, parties, events, crowd chats. What do you have going on? A ton of going on. So just recently, we launched the first Redis conference. So we acquired a company called Exceptional Cloud Services. We actually acquired the Redis brand. And at this show specifically, we launched support for Storm and Kafka, which is absolutely relevant to anybody doing any type of streaming from social platforms or otherwise. And we have gone from a very small team focused on data services to a business unit of 350 people, including support, marketing, over what time frame? In about a year. Wow, that's fast. Yeah, so I mean, really just exploding and adopting open source platforms. We have Elasticsearch coming down the pipe next month. And then a couple months, a Cassandra offering. So we're just going to keep going. We're just going to resize our Elasticsearch cluster this morning as we speak actually. You know, Elasticsearch, this is the new, this is the developer environment they want. They want the integrated stack. Yeah, absolutely. And we're seeing huge adoption. In fact, we're getting a line out the door, basically, to do that, saying, how can we be in the preview? How can we get there? And really just a good, easy, scalable tie-in to a lot of other data platforms. All right, well, let us know what's going on. Let's keep in touch. Sean Anderson, data services, booming business unit at Rackspace, really relevant. It's in cloud as DevOps. And this is the future of analytics, cloud analytics coming together. This is theCUBE. We'll be right back after this short break.