 Okay, we're live here in Silicon Valley, actually in San Jose, California. This is theCUBE, SiliconANGLE.tv's flagship telecast, where we go out to the events and extract the signal from the noise, really around the future of technology and innovation, all tracked on SiliconANGLE.com and wikibon.org, the research team for our group. I'm John Furrier, the founder of SiliconANGLE.com, and we are here at Hadoop Summit where Hortonworks, the recently spinned out company from Yahoo, venture back by Benchmark Capital and a slew of other tier one investors, are launching and putting a stake in the ground around their version and their approach to extending the ecosystem of Hadoop. Hadoop is the big data Apache software that is driving massive change in innovation. If you're watching this, you know that SiliconANGLE.com and wikibon.org have been tracking big data extensively for multiple years, going back now four years. We've been tracking big data. Back when Amar Awadala was pre-starting up Cloudera and our relationship with Cloudera and having an office in their office for a year and a half, now we've moved out, have our own office at SiliconANGLE. And we've been covering big data and we are really excited to bring you a great lineup today. Joining with my co-host, Jeff Kelly. Jeff, welcome as a co-host. Good to be here, John. You are the top analyst in big data right now. You put out the first market sizing report, hence then, followed suit. And still, even in New York City last, yesterday, people are still thinking that the numbers are still smaller and that they actually could be bigger. So I want you to give us an update on your recent post, research you posted, and just where we are in big data. Sure. Well, most recent post, research we just posted yesterday actually, to coincide here with the summit, is all about Hadoop as an enterprise-ready platform. And are we there yet? And that's been one of the big concerns among the enterprise. Is it scalable, secure? Is it secure enough? Is it easy enough to use, deploy, implement? So those are some of the issues that enterprises are concerned about before they're willing to launch mission-critical applications and workloads on top of Hadoop. So our recent report talks about the recent advancements in kind of the four critical areas that we believe Hadoop needs to improve to really become enterprise-ready. And that's all around high availability, stability of the platform, as I mentioned, the ability to easily interact with the platform. Security, of course, is critical when it comes to data management. You can't, in this day and age, not have a solid security layer in your data management platform. It's just not going to fly in the enterprise. So these are some of the things we've been working on and doing some research around. And each of the vendors in this space, important works here, and also Cloudera, MapR, they're all doing some really interesting things around making Hadoop enterprise-ready. Well, we're excited to cover the show here on siliconangle.com, siliconangle.tv here, our flagship telecast, theCUBE, which is our anchor desk format. We call it the ESPN of tech, where we go out and talk to the tech athletes, CEOs, entrepreneurs, big company leaders, whoever's got something really important to say, no matter who they are, we want to talk to them, extract the signal from the noise, and share that in an open-source way with you. So theCUBE will be here. We also have a new addition, Jeff, to the program here. We now have Studio B. Studio B is our second element of theCUBE, where we have an on-demand ability for thought leaders to come check into theCUBE. If they can't get on theCUBE, if they don't... Or if they're nervous to be on theCUBE, because some people are a little nervous to be on theCUBE, they can come here and check in with video with our studio team and get on the record with theCUBE. So we are expanding our operations. We will be going to a 24-7 capability very shortly in the fall, and now expanding our studio capabilities. And I want to give a shout out to Fred Davis and his crew at Burr Studios, who are working with SiliconANGLE exclusively on this new operation. And so we're going to extend more capabilities because we want to bring as much action to you and go to where the stories are and have the stories come into us and share that with you. This is a really, really big Hadoop Summit, Jeff, because what's happened in the marketplace right now is the Hadoop community, which started out as a very close-knit kernel, has expanded very rapidly with the first commercial venture with Cloudera getting venture backed. And then last year, a year ago, Hadoop Summit here, Yahoo spun out Ortonworks. A lot of bunch ex-Yahoo guys who actually helped build Hadoop from scratch created another commercial organization at the community level to extend out that capability. Since then, we've seen EMC, we've seen HP, IBM, every other big company come in and try to play with Hadoop, trying to figure out do they fork it, do they take it over, do they align with it? And then a slew of other startups around the ecosystem. So the ecosystem is really exploding and I'm expecting that this year that the ecosystem is going to dwarf even further. So it's expanding, the dynamics of the ecosystem is changing and also the market's changing. And you can see from the tracks here. So if you look at what the tracks are and the sessions, they have a bunch of different sessions. Obviously, 101 Hadoop for the people putting their toe in the water, the future of Hadoop, so for the people who are coding and contributing, where's it going to give some roadmap around that. Enterprise data architecture, this is where the money is and that's going to be where the action is and that's going to be where we're going to see some interface between the big companies and the existing scaled marketplace solutions with the emerging ecosystem and we'll talk about Ortonworks and Cloud Air and specifically around that new dynamic. Operations, DevOps. DevOps is Hadoop. Hadoop is DevOps. Developers love this. There's an operational aspect to batch meet real time. So a real operational aspect and we saw that clearly at the eight space conference when Facebook was actually really sharing some very excellent operational efficiencies with Hadoop. So operations are key in this. You can see impact there and that's going to have an impact on data warehouse and business intelligence. We'll get to all that on theCUBE and finally data science and applications and analytics. This is where the bread and butter is. The data science is the personnel side of the equation that's the labor and then the analytics is ultimately going to be the end product of business value. So this is a huge, robust show. It's very alpha geek. It's an elite force. This is not a suit show. Not a lot of suits here. Although there will be VCs and we'll be covering it but this is a tech show. So what do you take of this? Obviously what's your angle on this? How do you, as an analyst, put this together? Yeah, there's a lot to take in. I mean, there's so many different aspects of this show as you mentioned, as you ran through. I think for me, as I mentioned, some of the things we're looking at and what I'm really looking at is talking to some end users, some potential end users. Companies who are considering Hadoop get their opinion. Is Hadoop Enterprise ready? What is the feeling within the enterprise around deploying a technology that's still emerging really? Because ultimately it's going to be the practitioners, those companies that take advantage of big data. Just as much, if not perhaps more, than some of the big data vendors that are really going to reap the rewards here. It's the companies that get started earlier, craft their big data strategies now. They're going to be the ones that in five, 10, 15, 20 years you're going to see just thriving. So I'm really interested in talking to practitioners, talking to end users to kind of gauge where, how the enterprise feels about adopting Hadoop from integration with existing systems standpoint, but also from an analytics standpoint and that whole cultural piece. Analytics, big data analytics is more than just technology. It's also that cultural piece. You often hear the term, I don't love it, but a data driven organization. And that's just as much a way of thinking as it is a technology. So that's what I'm interested in. Okay, so if you're watching this right now, please spread the word. You can reach us by tweeting to us at Furrier, at F-U-R-R-I-E-R. And you can go at Silicon Angle or at Wikibon. And you can also tweet at Jeff Kelly, but go ahead and tell him your Twitter handle. Yeah, it's Jeffery F. Kelly. It's a long one. Jeff Kelly, unfortunately, is a pretty common name. So those simple Twitter handles are taken, but if you tweet at Wikibon, we'll get to your role. We'll see it. So tweet your questions and ask us what's on your mind. We have a slew of questions. I wrote a post just this morning that we wrote on the plane last night. I went back from New York data center conference about what we're going to do today. So go to siliconangle.com and read my analysis. The title of the post is Hortonworks Plays It Smart by partnering with mainstream adoption. That's going to be the theme of the show, the ecosystem. Now the founder of Hortonworks already retweeted that with some commentary. So this is where all the action is. And I have a slew of things that we're going to be asking on that post. So follow that post if you want to look at some of the things we're talking about. But really, from my perspective, this is about the ecosystem. This is about Hadoop intersecting with mainstream adoption. The hype cycle is moving into reality. Big data is now end to end value proposition from really narrow military operations to all the way across the board to retail. And everything in between has a big data aspect. So with that, I wanted to ask you, is your forecast wrong on the market sizing? I mean, you had a pretty good number. But we had the market set around five billion this year and growing to about over 50 billion by 2017. We might have been, I won't say wrong, we perhaps were a little bit low. Understated. But these things are evolving, these models that we use, just like practitioners of big data analytics. So actually from the beginning, your usual co-host and my boss, Dave Vellante, said from the start, Jeff, I think you got the number a little bit low, or maybe a lot low. So yeah, I mean that number could potentially go up significantly. Low or relative to the market, but actually higher than everybody else. For the most part, yes, that's true. IDC, Gardner, they all got it wrong big time. Yeah, well, I don't think, I don't think, you need to take a big picture, look at the market from not just the software, not just the distribution vendors, but also the application vendors, the impact on traditional IT vendors, and of course services. Services is huge in the big data landscape. So when you take all that into consideration, and that's how we came up with our forecast, I definitely think it can grow a lot bigger than the 50 billion. 100 billion is not out of the question or even higher. So we didn't get it wrong. We simply, we're evolving our model. Okay, so we're here at Hadoop Summit. This is live, this is theCUBE. We're expecting a great lineup. We're going to have the CEO of Hortonworks on this morning. We're going to have the founders of Hortonworks. We're going to have a lot of the industry players. We're going to have Cloudera on. All the fault leaders and all the luminaries. This is an absolute tech conference. This is where the alpha geeks, the leaders in the Hadoop ecosystem around open source is going to be here. It's going to be very much development focused. All the sessions are packed. It's sold out over 2,500 people are here. And you're going to see the true tech athletes here at Hadoop Summit because this is where the action is. The market is on fire. We're going to cover it like a blanket. And my expectation is that we're going to hear a lot of discussions around high availability, the name node question around name node, and also the analytics. The number one question that I'm hearing in the field talking to customers and developers is how the hell do I get the analytics out of HBase and the databases? Because there's no question unstructured databases are going to be interfacing with structured databases. That genie's out of the bottle. That's a done deal. Question is how do we get it from there? So what's your take, Jeff, on that dynamic? Well, I mean, look, analytics and applications is really where the business value is, as you mentioned earlier. The end result of big data is being able to make better decisions based on better insights. And that's all about analytics. So, you know, there are a couple different models or approaches to bringing the structured and unstructured data together. And you said it's critical because, I mean, that's one of the key value propositions of big data is that it allows you to bring all your data together and for a comprehensive view to run analytics on all your data without having to sample or without missing key data sources. So if you can't bring it all together in a seamless way, really one of the major value propositions of big data goes by the wayside. So in terms of how technically we're going to do that, you know, I think when we saw Hadoop start to develop, we were seeing, you know, connectors, basically, is how people were kind of integrating their structured world of databases and data warehouses with the Hadoop world in H-PACE. And bringing together, maybe crunching a lot of data in Hadoop, porting over some of that results into a traditional structured environment and running some advanced analytics there. But, you know, since then, we've seen, you know, a number of different vendors have built their custom connectors to make that process easier. But the question is that a long-term, is that a viable long-term way to approach this problem? And, you know, it remains to be seen. I mean, there's a company actually up where we're based, Wikibon's based out in Massachusetts called HADAPT, which is attempting to build kind of a robust SQL, like capabilities into their platform, into the Hadoop platform, essentially. So Hadoop can handle both of those worlds as opposed to having moving data back and forth. It's an interesting approach, you know, there's a long way to go there, but it's definitely an interesting approach because, you know, moving data around, I mean, that's one of the no-no's of big data, right? You bring the computer, you bring the analytics to where the data is. So, you know, there's a question mark right now. You know, right now the connectors is kind of the way to go. We'll see how that evolves, though. So we're here at Hadoop Summit. We're going to talk to all the top leaders. Jeff Kelly, the analyst at wikibon.org, our research team with the SiliconANGLE team is on the ground co-hosting with me. Dave Vellante is in Boston. We were just broadcasting live at Dell Storage Forum. So shout out to Dave Vellante and Mark Risen Hopkins and Mike Jones, Mick Jones in Boston. Guys, wish you were here. We are on a summer tour. This is the theCUBE. We go to all the events. This is our event season. And this is going to put a cap on our first leg of the summer tour, Jeff. We've got Google I.O. around the corner. We've got VMworld coming up and in between a lot of other different events on siliconANGLE.com, siliconANGLE.tv. Obviously, we've got wikibon.org, which is our research community. We'll be doing a lot of peer insights from the Boston Cube location in your office. So stay tuned for all the action here on siliconANGLE.com at Hadoop Summit. We're going to have all the live interviews and wrap it up in a short break.