 Okay, we're back, this is Dave Vellante, and this is day two of EMC World. This is theCUBE, SiliconANGLE.tv's continuous coverage, where we go into the tech events, we bring you the brightest stars, that we like to extract a signal from the noise, and we have a good friend of theCUBE, Bill Schmarzo, who we've anointed the dean of Big Data. I don't know if we started, but we kept it going, Bill. Welcome to theCUBE again. Well, I'll definitely give you guys credit for it. Yeah, thank you. But I think John Furrier tweeted out that week we were at Strata. Yeah, it was at Strata, yes. I came up with it. I think it was Ed Dumbill or Alistair Crowell, and we ran with it, and that was a lot of fun. That was a great event, wasn't it? It was a great event. That's always a great event. It's just, it's so invigorating to be there, because you've got such a vibrant audience who is so interested in conquering the world. And they don't know what's in front of them yet in some cases, and they're just balls out just trying to make things happen. I love it. And John, as we were talking off cameras, at the H-Base conference today, and I've not been, but it supposedly sold out. I think we're going to see a similar sort of attendee, a lot of alpha geeks, a lot of business people trying to figure out what's going on, and H-Base is at the heart of that whole thing. Although, there's alternatives, right? It's not the only one out there, isn't it? It's true, right? It's very true, yeah. Yeah, so tell me what's going on at EMC World for you. How are you spending your time? So I'm spending my time in two areas. One, I'm meeting with analysts, sort of sharing what I'm hearing, the noise out in the marketplace, which is always interesting to me to hear their perspectives back. And I don't like being a talking head. I want to hear what other people are thinking, and in particular, what the analysts are hearing out there as far as what are the challenges in the big data space, and how does that correlate to, what I'm seeing as I talk to customers out there. And the second activity I'm doing, I spend a lot of time talking to customers out here, quite a few of them out here visiting, getting ready to go to the data science summit tomorrow. So it's always great to talk then to figure out, why are you here? Are you trying to learn? Where are you on the roadmap towards, on this big data journey? So what are the analysts telling you? What are they asking you, and what are they seeing? Summarize that. So here's how I'd classify it. If they're analysts that come from the traditional BI data warehouse space, the audiences they talk to are still kind of sitting back waiting. There's a lot of uncertainty regarding how some of these new tools like the HBase, the Hive, the Hadoop, MapReduce, how that plays within my existing BI architecture. So there's that audience, but if I talk to analysts, come from more of the data science sort of perspective, or big data. I mean, they're all about new data sources, new application, new technology, new opportunities. I see those analysts much more enthused about what's ahead of them. And no digging the BI data, where else guys, that's my world too. But it seems to be a lot more apprehension by those folks, and I figure out, in many cases, they're customers, I'm not even quite certain what big data is yet. And I kind of find that surprising. Yeah, I think, you know, I had just said Chuck Hollis on it. We weren't able, we didn't have enough time to get into this, but he wrote a blog piece way back when, maybe it was right around the time you guys got it to acquire Green Plum. Basically saying that big data's an opportunity, you know, not a liability. Yeah. And so he was one of the first that sort of started talking about that way. But people look at data growth and the problem of volume of data as a problem, an issue that has to be dealt with and managed. I think that new group of people looks at us, wow, how can we get our hands on this stuff and monetize it and make some money out of it? You're exactly right, Dave. Exactly, I think there is a group of people who see that data, within that data, is opportunities to transform from being a market follower to, you know, a first mover advantage for gaining insights on the markets, products, customers, competitors that they can leverage to drive differentiation in their business problems. I went to the breakout session you had at the EMC CIO Forum in October, and we saw that there, right? There was a lot of angst, and then there were some people in the audience, it was a minority, but there were some people in the audience, and of course that was one of them, very excited about this, you know, because I didn't have to deal with the traditional IT mess, you know, it was all a green field for us. So we just did a survey on Wikibon, or the Wikibon practitioner community, which is a lot of traditional IT people, and we asked them what's the biggest big data problem, and number one was figuring out how to monetize and get value out of the data. So that makes a lot of sense to me. All the rest of the problems that, you know, that they cited or we cited that we had them rate and rank, very much sort of vanilla. It was all, yeah, yeah, yeah, yes, yes, you know, sort of, sort of, sort of. And I think the reality is they don't know yet. They're not there yet, you know. They're not there yet, and my frustration is they're not trying in some cases. And you know, and that's easy for me to say because, you know, I'm a consultant, but I just see people, some companies are just so hesitant about making the wrong move that they're not focused on even trying to figure out what sort of right move there is out there. They're more focused, I'm not sure if I can, how I'm supposed to say this, but they're more focused on making mistakes and on making game-changing moves. And I think that the new big data culture has got to be a let's try it culture and see what happens. Yeah, you're going to get it wrong. Oh, no doubt. I mean, I had interest. Somebody said, well, I had a customer last week tell me, I'm chartered to develop a three-year plan for big data. And I said, that's a wasted effort. I said, why would you? I said, you got to figure out where to start first. And then you'll get there, but if that's your first move is to figure out a three-year plan, you're going to get nowhere. Yeah, so then the other interesting thing that came out in the survey was we asked people, what initiatives are you going to hire outside services for this year in 2012? And we had a lot of cloud deployment and cloud management. Big data strategy was like two or three, right up the top. So it clearly says, we don't know what we're doing. We got to get our strategy together. Big data deployment, way, way down. Hadoop deployments, way, way down. Not surprising. So okay, so customers are basically saying you the same thing, hey, we need to do a three-year plan. In three years, the whole thing is going to be just, three years is way too long on the horizon. Well, look what's happening in the technology underneath all this stuff. I mean, Google comes out and announces a big query, right? Well, what does that mean? Well, no one knows yet, but it's probably going to have an impact. And so there's no way, if you were developing a three-year plan and it started it three months ago, you missed all of that and you got to go back and re-rig your three-year plan, I guess. So let's say the CEO is on the plane, read some article or whatever. He sees some video from theCUBE. He says, hey, this big data thing is really interesting to me. In my gut, I feel like there's something there. It's not just a bunch of marketing hype. So what's our big data strategy? So they go to the, whomever, maybe it's the CIO, maybe it's the head of sales and marketing. I'm not really sure who the right person is. It says, figure it out. So they come to you, hey Bill, you know a lot about this stuff. We need to get together and get our big data strategy together. Where do they start? What should they do? They really try to push people to stick their toe in the water. We've talked about how we run these vision workshops to help them understand sort of the realm of the possible. Not only for the IT organization, but also for the line of business. I mean, if you're trying to do customer retention programs, how can you leverage social media data to improve that effectiveness? If you're trying to improve your customer satisfaction, how does all this external data out there help you to get a better feel for sentiment on the services you're providing? So we try to get people to sort of start by doing something. Pick a business problem, understand how these new data sources and these new technologies like Hadoop, like R, like HBase and Hive, how you can use these things to try to glean out new insights about your customers, your products, your markets, your competitors that really can drive some business value. Start there, right? And as you go, evolve as you go along. One problem at a time, do the vision workshop, do an analytics lab, look to operationalize that, and just kind of build slowly on this process. You don't have to dive into the water head first because I don't think people know where the stumps are buried underneath the water. But there's no reason to stand on a sideline to figure out what's going to happen. You'll start wading into the water. One of the reasons I'm so excited about big data is it cuts across so many industries. Virtually every industry and every company is going to be affected by data. A lot of times, you know, people say, well, that's not our core company. That, you know, that internet trend, they really had browser, that's really not our core company. Mobile, that's not our core company. But data, should data be a core competency of every single company? You know, Joe had his keynote yesterday. I thought it had a really interesting slide. He said that the industry overall is transitioning from the application as a center with the data around it, to the data as a center with the applications around it. And I think that's what we're seeing is that people are starting to realize that the applications aren't quite as valuable as the data themselves. And the data that's important is the data that's about your nouns, your business. What are your nouns, your customers, your products, your partners, your markets, right? And the idea is that you basically focus in on building those nouns with outside data sources to really strengthen your insights on what's going on. So I actually think that there is a big transition from people who are focused on trying to optimize their applications, people are starting to realize, well, the data is what's important, and how I use that data, how I tease insights about data, how I use that data to make better, more frequent, more granular decisions. That's what it's all about. Let's talk about, you know, the other thing Joe said that really struck me, and of course he said it because we were on stage in there, and you saw us, John and I, with Maggie Burke and EMC-TV just beforehand, and Maggie asked John, what's the future about? And John said, the future's about real time. And Tucci basically pulled out what it was, and I'm giving a prize. Trust me, it's going to be a good one for the naming of the guy, but then he said, I'm going to predict real time is really where all the action's going to be. So I want to talk about real time, because you and I have talked about Hadoop a lot, Hadoop world, and everybody's Hadoop crazy, but Hadoop's batch. Hadoop's not real time, it's not real time. Is there a dissonance there? Well, Hadoop does what Hadoop does well, right? And I think if you try to think of Hadoop as, there are people who say, oh, big data equals Hadoop, right? And that's the first problem, is that you've got a richer set of tools out there now, not just Hadoop, but you have your traditional behind data warehouses. You have all these in-memory computing, like, you know, we've got gemfire and SQL fire from VMware and such. It's really putting together the right architecture that allows you to do what tools do best. So do what Hadoop does best and do, which is not real time, but the ability to process massive amounts of data, to look at unstructured data, to tease insights out of that, to, you know, read versus schema unwrite sort of stuff and agility that provides, use the right tool for the right things. And I think, by the way, that's part of why IT auditions are paralyzed, because it isn't like, oh, I just do Hadoop and I'm good. It's like, no folks, you've got to look at a whole wide range of tools and figure out which of these different technologies and tools are right for what you're trying to achieve. And by the way, if you don't know what you're trying to achieve, you're in trouble. Yeah, so back to the strategy, you know, point, is maybe start there. What's the objective, you know? Hey, hey, hey. What mountain do we want to climb? So is your scenario then, so Hadoop is this batch system and then you'll find the nuggets in Hadoop, let them sit out there and they're, you know, in their map-reduced state and then find the nuggets and bring them in to a data warehouse system? So there's two different ways that we see cases for how it's happening. So we're seeing a case where people are sort of just dump everything in Hadoop as it is, right? Like the quote, Hadoop is a new tape kind of thing. And then once it's in Hadoop, you can use that to take down structured data and pull structure out of that, right? So you can actually find structure there that may very well find its way into your data warehouse. It may find its way into your BI reports and dashboards. So it's new insights about customers' interests, about their passions and affiliations and associations that find its way into the data warehouse. So you could see that your data warehouse environment actually gets augmented by the insights Hadoop can tease out of that data using Hadoop. But it's also, the other spectrum or the other place is how am I fueling my analytic sandbox? And Hadoop is a wonderful vehicle for feeding data as needed on demand into that environment. So Hadoop can help support both environments, but it's not the complete answer for both environments. I mean, you wouldn't have an analytic sandbox unless you had statistical tools like R or SAS, right? And so you've got to basically think holistically about your architecture and not just about its Hadoop. Yeah, so when you talk about the analytical sandbox, are people actually setting them up and what do they look like? Oh my gosh, yes. It's actually, there are people who are doing that. They're dumping data into it. They're playing with it. They're building models. They're failing fast. They're going out and grabbing more data. This is kind of the advantage of having this Hadoop as your operational data store and your analytic sandbox over here is that I can pull data in. I can play with it, find stuff. Oh, don't have enough. I need more of this. I need less of this. And go back and forth and grab data here without banging against your data warehouse, right? Without dragging down before into data warehouse. And then when you find insights here, they may find their way into the data warehouse. They may find their way into other operational systems. But then you start thinking about how I package that and operationalize it. Now Bill, you sit in the services side of the DMC organization and the whole theme of this event is transformation. It's IT plus business plus you. And IT, we see that as cloud. That's the big IT transformation that's going on. The business transformation we're saying is big data. And figure out how to get value out of it. And then the you is, all right, we're going to be a cloud architect. We can help you train to get data. You're going to be a data scientist. We can help you train to do that. Is that a reasonable way to look at sort of how you're messaging maps to what you're actually doing in services? Yeah, actually, we think about our engagements as a three-legged stool. We think about the first leg is the technology and the data as the first sort of the foundation. The second leg is the business initiatives. What are you trying to achieve? What business problem are you trying to solve? How are you trying to derive competitive advantage? And the third thing is the organization, which is not only about training skills, but also how are you building the right organization so your data scientists and your BI teams are collaborating around these business problems? How do you make certain that the data science organization can get data out of the data where else as needed? But it also has a way to publish insights back into your BI reports and dashboards and such. So we do think about those three legs as being the three sort of legs on the stool. Excellent, now you have this data science summit. They decided the summit coming up this week. Are you actively participating in that? I get to kick it off tonight. A little toast, kind of welcoming people. And then I get to sit back and listen and learn and talk to people. So I'm actually very interested in mingling with the people and trying to see what's going on because that's a self-selected audience, right? These are people who are already anointed themselves as data scientists and they're going to be further down the path than the people who I've met with, I did a keynote presentation at the Data Wall Institute about two weeks ago. Again, a very different audience of people who have different perspectives. So it's interesting to talk to two different audiences because the right solution for it is going to end up being somewhere in the middle there. Well, we were talking before about the sort of the schism between the traditional DWBI guys and the sort of emerging, hot, big data people. You're going to get a mix of those, I would imagine, at the Data Science Summit, right? I'd hope so. My fear is we're going to get mostly data scientists and not a lot of the BI data warehouse guys. That's my fear. Well, I mean, I'd be surprised because we're at EMC world, you'd think people, if in fact our survey data is right and people are concerned about the strategy, you'd think they'd pop in, see what's going on. It's obviously a hot topic. We'll see. I hope so. I guess last year, I'm trying to think, last year there was, it was mixed, right? I mean, very data scientists. It was data science focused. Yeah, there's not a lot of BI people kind of popping in here yet. Yeah, okay. I just think the BI market's been slow. They just want to move on this. And... Well, you know, the BI market, in my opinion, I'm just going to say you've been in that business for a long time, but it failed to meet its promises in terms of 360 degree of view of the business, real time, predictive analytics. It just didn't happen that way. And in a way, I feel like, you know, the whole compliance thing, the Enron blow up, saved that business for a period of time. And it became a rear view mirror looking business. And now the promise of big data, we'll see if we can live up to it, is what the vision that the BI world put forth 10 years ago. Yeah, the BI tool, or the BI marketplace, I've heard somebody say that yesterday's failed technology, but to a certain extent, it maybe was successful for what it was designed to do. It just was never had the capabilities to do real time predictive stuff. It never had the capabilities to do and look at unstructured data. All the things that are kind of hitting us now, the BI environment is okay for doing standardized reporting and dashboards and, you know, financial reporting. It's fantastic. It's a single version of the truth. It's fantastic single version of truth. As long as that truth is internal, you try to bring it outside data sources, your single view of the truth all of a sudden gets really muddled. And that's why the BI market sort of struggles. Yeah, and big data is fuzzy. Yeah, it's all a difference. Big data scares them. Big data scares them. Yeah, good. All right, Bill Smarzo, awesome seeing you again. Thanks very much for coming on theCUBE. Thanks for watching everybody. Keep it right there. We'll be right back. This is theCUBE, SiliconANGLE.tv's continuous coverage of EMC World Live in Las Vegas. Keep it right there.