 Okay, welcome back here, we're in Las Vegas, where all the action's happening in the Big Data Week. This is IBM's information on demand, I'm John Furrier, the founder of SiliconANGLE.com. I'm with Dave Vellante, my co-host, and this is theCUBE, SiliconANGLE's flagship program, and we're out for the events, and I'd like to see them from the noise. And our next guest is Rob Thomas, Vice President, Business Development, head of Big Data, you're running the Big Data Group, leading the acquisitions, leading the go-to market. You're in charge of the Big Data positioning for IBM? Yep, we're doing it, look, if you're not at this show, you're small data, and by definition, we are big data. So we're excited to be here, we're doing some fun stuff. We just had one, we had Stacey on from Vismo, one of your positions. So, first of all, tell us, the folks out there, what your group does within IBM, and then we'll go outside of IBM, how's the Big Data Group? Obviously it's hot, Big Part IBM's entire focus right now, Big Data. So information management, software business, focused on data infrastructure, basically making sense of all the plumbing in the organization, we are proud plumbers. Most people like to play up the stack, but we love being down in the weeds, and we're plumbers. And we've been on, I'd say, a long journey in Big Data. We've been in databases forever. We did the acquisition in Netiza a couple years ago, which kind of our first move into appliances, and done tremendous organic development around a Hadoop analytic product, a streaming product. So we've really kind of evolved our portfolio to cover all the components of Big Data. I think a lot of people say, Big Data equals Hadoop, that is not our view. Our view is there's a lot of components to Big Data, and that's really what we've been trying to build out. And the database market is on fire as well. You guys have all the right trends, you're on the right vector, you have all this technology. So it's not like you guys woke up one day and said, hey, we got to get into Big Data business. And not even like years, a couple years ago, for a long historic view, you had a holistic view of complex machines. We always used to joke that VMware is building the software mainframe. You guys have the mainframe with virtualization and Big Data. So that evolution is natural for IBM. So take us through when the Big Data really started to become key within IBM. Not necessarily the technology, but like to the business leaders within IBM. Was it five years ago, earlier? Was there a point in time you said, hey, you know this information management thing? Which was basically Big Data. Analytics, the cognitive acquisition. All that stuff, how far back does it go time wise? And tracing back to where we are now? So let's go back about seven years. At that time, we had a business called Information Management and it was DB2 and Informix. That was it. That wasn't really information management. That was a database business. That's when we first started talking about a vision that said, how do we stitch together all the pieces? And so we did the essential acquisition for ETL. We did Cognos for BI. We did FileNet for BPM and Content Management. And so we really started to build out the portfolio there. I'd say then, so that's what became IOD. That people wonder why is this conference called IOD? It was kind of when we made the move from being databases to being everything that surrounds a database. The next inflection point came, I'd put about it two and a half years ago and I give the IBM marketing team that came up with the Smarter Planet concept immense credit. I mean that was kind of the start of Big Data. We started to see examples. We were working with City of Stockholm on traffic, working with Vestas on wind turbines. These were things that were changing industries, changing economies, changing countries. That led to Smarter Planet and now we're in the middle of Big Data. And obviously we've had conversations with Anad Jhingran who's now moved on to startup world but when we came down to your office, even going back when I started doing podcasting, this is back in 2005, Enterprise Search evolved pretty much in decade was picked up by Oracle. So that whole search business became a part of an element of Big Data. How does that playing into all this? Obviously you need to find the data, right? It's almost a demo of it. But now you got data sets. You get diversity of data, graph databases, time series databases, Hadoop relational. So take us through why that piece is important and how does that fit into the picture? You know I wrote a blog post about a year ago where I said I thought 2012 would be the year of failures in Big Data. And the reason I said that was we'd done so much work with early adopters and we were starting to realize how difficult it is to actually get value out of these newer technologies. So we go through all these engagements and eventually started to have the insight that said we need a better way for clients to get started with Big Data. Inner Vivissimo. So I got to know the team from Vivissimo. Three guys that founded it from Carnegie Mellon, tremendously talented to team based in Pittsburgh. You said that. Tremendously talented team. But what they're really doing, their secret sauce is not enterprise search, albeit that's how they started in 98. Their secret sauce is they can get a client up and running on Big Data in a couple weeks. I don't think anybody else can do that. Most people can't even set up a Hadoop cluster in a company. But not just Big Data, I mean, leveraging existing data systems. Not necessarily, I mean I can throw a cluster up on Hadoop and start grabbing data and dumping stuff in and maybe playing with it in the sandbox. Their value is what I call data discovery. So you go to any client, when I go to a bank on Wall Street and insurance company, healthcare, number one thing I ask them is, do you even know what data you have? Most people don't. They say, well, I've got this system, that system, that system. I say, why don't we start with you understanding your data assets? That's what Vivissimo is. So Wikibon, Dave's team, and SiliconANG, Wikibon team, size of market, what was it, 50 billion? When we first, we were the first ones with the sizing. Yeah, but 2017. And we're doing a lot of work. Obviously we're going to be at Strada, I'm flying out tonight, Dave's going to be here tomorrow. I'm going to go set up the queue for our second run. It's a little bit different, Mark. You guys are animals. This is like the college game day of Dorks. It's like every week you're going to the next hot shit. Dorks and nerds are just a little bit different. That's why the East Coast word, I take offense with that. We're geeks. Yeah, we're geeks. No, not Dorks. We're cool, actually. You guys are doing a tremendous job. I'm really impressed. I remember, John, when you and I first met a couple of years ago, you were describing this vision. It's amazing, it's taking off. You guys go everywhere that's hot. You know, we were a big, we were a big, you know, we were a big data. It was interesting, because what I love about what you guys were doing is why I'm so enamored with IBM and proud to be kind of part of it, watching them from the seats. The chief seats is that we talked about big data, and when I was starting SiliconANGLE, even before I met Dave, and Dave has a shared vision, is that it's a big database. So our entire publication business is based on big data using project analytics, and you know the story. So we don't run into banner ads on our website. That's why we use date data for all that. But we're going to Strata Hadoop World, different markets. So you mentioned failure to big data, but you didn't factor in some other dynamics, which is the market is so thirsty for data. They want a drink from the fire hose of big data. They want solutions. They want proof of concept. So it's kind of a sandbox market for the past two years. Right? Okay, we'll buy that. But now it's kind of talking business value. So talk about what you think's going to happen at Strata upcoming, because the middleware market in the startup circuit is kind of on fire right now. The guys say, hey, let's build on top of H-Base, this need to automate. Are you going to Strata? I'll be there at the tail end, I think. You got to go. If I can get out of here. I had to ask to go, because you're like a kid in a kid's story. I know, I know, exactly. What's going to be the storylines at Strata for the emerging companies? You know, I think- The guys you're watching to buy, by the way. That's right. So you know, I think about, like I said, I thought this year would be the year of failure because early adopters went fast and it was difficult to get value. What I've seen the companies this year that I'm meeting with newer companies is in a way they've gotten over the chasm to some extent where they can deliver business value quickly. So the companies that can help a client get value fast are starting to make a difference. That's not where we were a year ago. So on the M&A side, I mean, you guys are public companies, so you really can't reveal the secret portfolio targets. Pretty much it's everybody. Just like EMC and HP. What do you look for in a company when you're looking on the M&A route? Just generically. I mean, you mentioned speed to value. Obviously, people up and running. Are there other characteristics beyond that that you look for to start up when you say, hey, you know, you actually have some white space to fill fast in product line. What do you look for? What are the key things, say, good team? Is it all the kind of startup cliches or is it something different with its IBM's perspective? So, you know, honestly, in acquisitions, the team is often more important than the technology. So that's number one, because our view is the technology is critical, but you have to have the team that can evolve it to keep it relevant. So that's number one. Two is we look for businesses that have been around for a while that have real traction. It's less about size of revenue. It's more about number of customers or people actually using your stuff. Because our view is if you've shown enough traction where you've got hundreds, 500,000 clients, that's something we know we can scale because there's enough people that have trust in it. Even if it's small revenue because it's a bunch of small transactions, that shows that you really have- Where IBM can scale up with it. So it's a scale machine. You guys look at IBM and saying, IBM big blue is a big scale machine. Can we plug this guy in, this company, and will it grow fast? Yeah, we're giving an example. So when we did Netiza, the day we bought Netiza, they had 1% of their revenue outside of North America. 1%. Oh, you must have loved that. So that was- Oh, you don't do any business in Africa. But that's what we do best. We're in over 180 countries so we can take that model and scale it. And now we're in Brazil, we're in Russia, we're in all parts of Europe, we're all over Asia. That's the kind of business that we can grab and we can scale. And the companies love it because the engineers at Netiza see that their technology's all over the world now. Can you talk about the Netiza acquisition? That's one of the questions that I wanted to ask you, Rob. I mean, they went first of the next gen, sort of data warehouse, data appliance companies. Why Netiza? You had to pick at the litter. Yeah. Why'd you go with Netiza? You know, I got to know Jit Saxena, who was the founder of Netiza, probably over six years ago now. We started working on some partnerships. Never quite got it worked out, but eventually they adopted the IBM hardware platform. And you know, if you get to know a guy like Jit, he understands clients incredibly well. He's been there and done that before. We got, you know, really started to believe their value proposition around simplicity, around time to value, get up and running in 24 hours. We saw clients doing it. We worked with them on a partnership. We saw these are the kind of guys that follow through on everything they say they're going to do. Just great people to work with. Got to know Jim Baum as he came in and took over as CEO and, you know, it's like everything else. You decide these are the kind of guys I want to work with. And then on top of that, they've got tremendous momentum in the market and a tremendous product. So it's actually a pretty easy decision. Yeah, okay. And then a couple of other trends that we've been tracking. There's a big theme this week at Strata around unification. You know, sequel, no sequel, structured and unstructured. What are you seeing there? Do you think it's a little bit ahead of the time? I mean, you saw Hedat made a big announcement. Hortonworks, a bunch of other folks are coming out this week. Cloudera's going to have an announcement. So a lot of action going on there. I presume the market's pulling that. What are you seeing there? It's hard to say. I almost think of the no sequel movement right now as it's kind of like the music business where bands get really hot for a year and then you don't hear from that band again. And you think of it, if we were decided to go do, you know, a deep partnership or an acquisition in no sequel a year ago, you'd probably go do something around Cassandra. Facebook's using Cassandra and that's all the talk, right? That's kind of gone this year, right? Now it's on to MongoDB and the new thing. To me, we're going to watch this market. We would like to see it settle down a bit. It's hard to chase the next big music act. And we'll have some of our own organic plays there as well. But for right now, we're a bit of a wait and see. Another trend we've been tracking is just this whole notion of, I mean, everybody's talking about Hadoop, making it enterprise ready, robust. Everybody talks about that. You guys do a good business there, but what about security in Hadoop? Generally and specifically at the database level, is that something where you see, you know, serious white space and opportunities for startups? No question. And I'm actually a little surprised. There's a couple of guys I've seen that are starting to do stuff around that. I think one reason it hasn't taken off yet is there's not enough clients running, I'd say, mission critical in production data sets right now in mass and Hadoop. But that's going to change as really next year. Looks government. It's coming, yeah, it's coming. You know, media financial services, healthcare. So I think that's a tremendous opportunity. The Acumulo project is sort of, you know, that's what that's all about. And then, I don't know if there's other activity out there that you're talking about. I'll tell you what model I think will take off, and this is one reason I got so interested in Venissimo. Not to get too techy, but they ingest the security profile of the systems you already have. So they don't create their own security model. They don't require you to go build one. They basically say, if I'm allowed to access stuff on this system, but not this system, but I can't on that one. When you build the index, it automatically, when it gives me the UI, it only gives me the data that's available to me. It takes the security policy with them. Yeah, it takes the security. And I think, you know, as I look at big enterprises, if you can ingest the security policy, it's already, these guys have already spent millions of dollars to get that in, right? That's a hell of a lot easier to trust it. Yeah, than trying to go in and say, here's a new way to do it. Whether or not it's the best or not, it's theirs. So obviously analytics is hot for you guys, which is, you know, not surprising with IBM, but I want to ask a little bit different question on that point. Everything comes back down, all of our CUBE conversations are on big data. Back to the people conversations. You mentioned team and startup, the people in the organizations, the peoples of the barriers to getting the things done. It's a people, awareness, education, et cetera, et cetera, it goes down. The people is the central variable in all things that's happening around big data. So talk about data scientists first. And that role, how real is that from a height versus reality? And what's that future position or positions look like? And then other people related things that you've seen in big data. As a opportunity and a challenge. Yeah. You know, I think data scientists is a clever term. I think in a way it's a term for what we've always done is we've always had people in the organization that just study data, look for trends, look for needles in the haystack. So to some extent I don't think there's a change there. What has changed is the tools have changed. So it's a lot easier to be effective in that type of role. Two is the amount of data has changed, which means there's more to look for. So I do think that role will take increase in importance, but I'm not too sure there's something so unique about that role in and of itself. It's just a new set of tools that we've touched on. It moves up the stack in terms of user and analyst now could be a data scientist because they don't have to be a total quant job or programmer. That's right. It's more like the tools, the ease of analysis. That's right. Maybe there. You're right. People who know the business problem may not be close to a solution, but actually engineer a feature. And then it'll have to be a PhD. In a way, yeah. In a way you dumb it down for people like me, right? Even I could be a data scientist, at least in a bad company. Okay, so back to strata. So what do you expect to see there in that ecosystem? Obviously the ads from traction, what do you expect to see at strata this year? You know, so we've got over 300 partners that we've enabled on our platform in the last year and a half. So tremendous progress from an ecosystem perspective, ranging from visualization tools, a little bit on the security side, a little bit on the file management, a little bit on cluster management. I think we'll see more and more of that type of thing. More, I'd say, creative use cases. I think what I would really hope to see is companies that are delivering vertical specific use cases with Hadoop under the covers, or with streaming data under the covers. I've seen some of those examples. I like to be a finder. And they've seen some of those examples and partners we've worked with, but that hasn't taken off yet. Yeah, so there's a green field opportunity. Yeah. And those are the new questions that are answered. That's going to create those kinds of new companies, right? Yeah, that's right. And Mills talked about that the other day, that those vertical specific use cases is really where, you guys have got very strong verticals. And they're different. I mean, the horizontal stuff is maybe not as applicable out of the box. We need to see that ROI in these vertical cases. The key thing that you're right is to get ROI for the business guy in any given company so they understand why is everybody talking about this thing? I got to ask this question because being an old IBM, or I'm always interested in IBM's history, I was a co-op student, so not really a true IBM back in the day, but I'm confused by the whole storage group. Okay. Because like we did IBM Ed, which was a fantastic event and Jeff Jonas on, and you had a little bit of Tivoli there, you had the storage group, you had some other factions, but we're not hearing anything about storage here. So obviously Big Data has storage somewhere. Yeah. So how does storage fit into the equation? The storage group, like the drives, the rates, all that stuff, what does that fit in? I think we're actually at the edge of, I'd say a major innovation in storage. So I think storage will change dramatically in the next decade. I mean, for some reason that you see some of what you do is traditionally IBM storage was in the hardware business. It's kind of like it was the database business back in the day. It's like, oh, you guys got some disks and you're pumping out through OEM deals or whatever, right? That's right. Now it's like, wait a minute, strategic asset. Is that similar? Yeah. And I think what we're starting to do now is I'd say bigger cooperation between the software side of IBM and the storage side. So you guys have probably heard some about what we're doing around defensive, disposal, intelligent archiving. That is essentially bringing software intelligence on top of storage, doing some of that logic actually in the storage arrays. It delivers an interesting- I was talking to an IBM partner and we were trying to flesh out this whole small, medium-sized enterprise value proposition. Because with cloud, you're going to start to have stuff in the cloud. And I said the storage group, which targets that area, they said big data doesn't mean anything to the storage group because they're all bunch of hardware guys. So talk about this dynamic because we're seeing this year a software-defined virtualization, the network layer, is seeing virtualization and software be a key differentiator in the hardware business. So is that a false statement? That's just kind of like, that's just the way it is. Are the storage guys going to stay hardware? Will they become more software? And what does that transformation look like? I would call it a blending of the stack, right? Traditionally, software and storage to some extent have been different worlds. But with some of the examples you use, you're going to start to see a tighter linkage. I will tell you, the one thing that I think a lot of people miss about value proposition around big data is the place that you can, the clients that save a lot of money is actually on storage. You can move off expensive. So big data is relevant to the storage group? Yeah, no question. I mean, unless it becomes an opportunity, it's going to be an enormous threat. I'll give you an example. So I talked in our big data session today about some work we're doing with nice systems. Big ISV out of Israel, they do call center software, that type of stuff. They've traditionally run on databases, high end storage, you know, pretty expensive stuff to deploy. Next generation, they're moving to their application on top of big insights, leveraging HBase. That moves you to a commodity storage environment. Let's talk about HBase, because we love HBase, our products built on HBase, the one we're using. And as we want to scale, it's great for what we use it for, but when you talk about production systems, bulletproof, real, major leagues, you can run HBase and run stuff within IBM. Explain that to folks what that means. So that the communities don't think it's an extremely exclusive situation. It's not, I mean, look, the requirement that we've seen with some ISVs we worked with, one I just mentioned, another one that we'll announce soon, is when they're moving mission critical apps and they want to run it on big data, they can't just do it on, you know, Hadoop and MapRedoos. So they need some type of way to really manage the data inside of Hadoop, right? So every time the requirement keeps coming up as HBase. And, frankly, that creates a number of complications in terms of, you know, can you persist data, eventual consistency, all the stuff that you can imagine? That's where we put a lot of our engineering time, is actually making that stuff ready for prime time. Ready for the device. So you want to eliminate the complexity by putting a hard and top on HBase. Hey, let HBase go crazy. Yeah. Just move it up to the next level. That's right. That's right. Cool. I was talking with the Hadoop community. We're seeing where there's so much opportunity and there's so much beach head for the big guys and the new guys, right? So, you know, I'm envisioning for the entrepreneurs out there. Rob is going to be, you know, he's got his checklist. He's checking it twice or around the holiday times. So, you know, I'm sure every company's a target at this point. Final question is what's the vision for you 10 years out? Okay, knowing what you know with an IBM, you've been on this nice trajectory, and you have a lot of, you know, good trajectory that you've investments over years, so you're in a good spot, well ahead of the curve. What's the next five years, 10 years, look like from a vision standpoint? So, let me give you a prediction first and I'll give you vision. So, I joined IBM in 99. At that time, there was all this talk about e-business. You guys probably remember the e-business. Yeah, yeah, yeah. Good campaign. And I remember seeing guys on CNBC, okay, are you an e-business? Are you not an e-business? Wouldn't you be an e-business? Everybody talked about that. You don't hear that today, right? Nobody's asked you if you're an e-business, like they assume. And my point is that e-business became the default, and so the term went away. I think that will happen with big data. I would be shocked if we were sitting here talking about big data in three years or five years, because it will become so fundamental to, it will basically be synonymous with IT. Think about cloud. We're not talking about cloud much anymore. Yeah, it's already started. So, I think, so my one prediction- I'm not sure cloud adoption is going to be as big as big data, but I think it's an element of cloud. Well, I do too. Well, what's the saying that big data gives cloud something to do? That's right. That's right. Rob Thomas, Vice President of Business Development, heads up the big data group, obviously a center stage here, information on demand, the legacy, the DNA, the embryo, back in the days, evolved from databases to information, e-business, now big data. Congratulations on all your success. This is theCUBE. We'll be right back with a wrap up right after this short break.