 At Big Data SV 2014 is brought to you by headline sponsors WAN Disco. We make Hadoop invincible and Actian, accelerating Big Data 2.0. Okay, we're back here live in Silicon Valley. This is SiliconANGLE and Wikibon's theCUBE, our flagship program where we go out to the events and extract the signal from the noise. I'm John Furrier, the founder of SiliconANGLE. I'm joined by co-host Dave Vellante, co-founder of wikibon.org. We're here live to talk about Big Data with all the action in Silicon Valley that's happening right here at the Hilton and behind us at the Santa Clara Convention Center is the Stratoconference. A lot of Big Data happening, a lot of business being done, a lot of tech being talked about. But of course we have theCUBE which was extracting the signal from the noise with all the best tech athletes. And our next guest, we're excited to have legend Andre Bouvere on theCUBE. Thank you very much for coming. Board member of Actian, industry legend, been around the block more times than us and has more war stories. So we're going to dig into great conversation. Pleasure to have you on theCUBE. Well, thank you very much for the invitation. Love to hear the story. We were talking before we came on live about some of the experiences around how Larry Alice has started Oracle and Big Data. I got, we always ask the folks would come on, have you seen this movie before? You know, we've seen, you've seen many movies and cycles of innovation. How would you compare this innovation cycle that we're in now compared to other ones that you've seen? Well, I guess I've been fortunate enough in life. I started in this business in 1976 by joining IBM. Actually, I was a microcoder. I used to write microcode for major frame assist for instructions for MVS. So I saw that I landed at IBM at a fortunate time when the customers were no longer bringing their data to IBM to get their data processed and then going back to their office with the results. IBM started to say, well, geez, why don't we bring the computers to you? And IBM started to sell computers. That's when computers got down through a relatively affordable numbers. What I guess I could share with you back then. A 5.5 machine with 16 megs was roughly 5.5 million bucks. That's just a CPU. First example of function shipping. Yeah, but I obviously had a value because customers, when I was there, we could sell every single copy that we could build. We couldn't build them fast enough. That's when customers decided that there was value to automating mundane processes. So the first thing that they did is they brought these computers on board at home in their data centers. And the first thing they automated was mundane tasks but needed to be done, payrolls, AP, AR, and so on. And that's why it was called data processing. Hence the title of data processing manager. And as the customers, and then I lived through when as the customers started to pick up data, they said, hey, there's some value in that data. What about if we were to make decisions based on that data? So being Canadian, I left and I joined a small struggling Canadian startup at the time called Cognos. This is 1989 and it was struggling. We had about nine months worth of cash at hand. And so we decided that we did a joint project with Procter & Gamble. And we built a product called impromptu and power play which is basically built on Microsoft Windows because Bomber had worked for Procter & Gamble. So Procter & Gamble loved Microsoft software. And at that time I was saying, Jesus, Windows, read an operating system, real men, run MBS. But I certainly learned that there was a change. So there was a change of basically a client server. So we were extracting data from mainframes, moving them on to a Microsoft platform. And we were giving clients, and when I say product managers in the sense of Procter & Gamble, these are people that either manage a toothpaste or a product line. And so they were able to drill down and extrapolate and envision what the data, based on this, it's okay, we'll wire sales down to Brazil. Well, let's drill down. Let's look at it as specific to a date, to a specific time of year or a specific geographical area of Brazil, right? We're doing well in Sao Paulo, not well in Rio, right? So they were able to make decisions. So visualization, that's why I kind of laughed in a hair of a tableau and quickly, that was 1989. We didn't call it BI, didn't exist, okay? We call them information systems, right? EIS, in fact, executive information systems. And so, we learned quickly that there's value to be extrapolated by going into the data that you've accumulated within a silo, meaning your own enterprise, and then making business decisions. But again, you're still dependent on that human being able to extrapolate what they're visualizing, and with their own biases, kind of putting an experience, putting the next dollar on the board, which could or cannot be accurate sometimes. And so, then I moved on to predictive analytics. And so, this is really where I had a lot of enjoyment is, and eventually became the president of Staff Institute, which is the largest privately held software company in the world in predictive analytics, three billion in revenue so far in the last year. And so, I really got to see the value that customers will pay you a lot of more money for an ugly looking green bar report, but that is fairly accurate in predicting what tomorrow's gonna look like, then giving them a nice three dimensional color graphics of how much money they lost yesterday. And so, I really, I think of all the jobs I've had, I think this is the one that was really, really interesting because it's amazing if you can build models that with a high level of predictability, allows customers to predict what the future will look like. And so, therefore, you take all the human element out and everything, you're putting in weather information, you're putting in basically financial information, economics information, and so on, so that retailers, insurance companies, and so on. I'll leave it nameless, but there was a large insurance company I used to deal with and they would sit there and they would do models and say, where do we think we're gonna get hit in the golf? Where do we think we're gonna get hit on the East Coast? Where do we think Tornado Alley's gonna look like this year? And based on that, the chief risk officer would sit there and say, okay, the following zip goes, no more, we're not writing any new policies, and by the way, people who don't make their payments, following zip goes, cut them off. And then next year we'll revisit where we want to ensure. Andre, I want to ask you a question. First of all, this is awesome. I can just keep the cameras rolling all night long. But I got to ask you an interesting question that we don't usually get on theCUBE, given your leadership and what you've done and what you're working on now, so the acting we'll get to in a minute, but because it's a great corporate development maneuver, Dave and I were commenting in New York about that play. But I want to ask you about leadership. In a time of change, we're in now, the winds are shifting and the money's on the table. And now it's, everyone's racing to swim out to the barge and get that cash, right? So get that value. So I want to ask you about leadership, right? What do you think of the leaders out there right now, the old captains of industry, like the Joe Tucci's, the John Chambers, to the new upstarts? And what does it take to be a good leader? If you can comment about Tucci and Chambers, that'd be great. Right, so they're two very different individuals. And funny enough, they both are somewhat related because if you go back in history, I actually worked for Joe Tucci and Dave Goulden when we took Wang out of Chapter 11 and had a legal counsel actually for, Wang was John Chambers. And so obviously they went their separate directions. I've got a lot of respect for both. What Joe has realized, I think, is he's forming these spin-off companies because he recognizes that if he keeps it within a big, monolithic company is where you've got to make a number. I mean, Wall Street somewhat creates their own problem that he can only do so much as a public company. And so by spinning off some of these assets and doing joint ventures like he did with GE, right? Those are great, those are great moves. It stimulates growth. You like those moves. I like those moves. And so I'm a big fan of what Joe's done. Cisco is a different model, right? Where they've done some acquisitions like WebEx and they've done some related acquisitions but they're not, in my view, not as aggressive. But again, you got to give it to John. He's done a great job with Cisco. One could argue it's always easier on this side of the table to criticize people and say, geez, but I think he could have grown at a faster rate. But I think, I'm still a big fan. I'm a shareholder in Cisco and have been since I met John back in 1994. You mentioned Larry Ellison. What do you think about Mark Benioff's move out of Oracle? What do you think about his moves? Well, again, ironic, Mark used to work for me, had a work call. And so it's a small world, so Mark was pretty wild at the time. Tom Siebel, too? No, no, Tom was down at the time we got there. But Mark was a very clever guy, very creative. And so that's the polite term for saying he was kind of hard to manage at time, but a very bright guy. Eagles are hard to nail down, right? I mean, hold down. Yeah, but he did, I will, and I've said that to him many times, is he's done more with his company than I would have ever thought he could have gone. And I think, frankly, I think if you were to talk to him privately, I think he would admit that it's gone much further than he thought he was. But the other thing I think that surprises me of Mark, and again, I'm not saying it in a demeaning fashion. But it's really been pleasant for me, and we gratifying to see how he's matured as an individual because you've got him in credit, he knows how to hire. And he's done a great job of hiring very, very good talent, and he's got a good sense of where the industry is going. And now he's got his challenges as well. His platform is what, 10 years old now, right? So what was very avant-garde at the time, he's now starting to live with same-of-the-same problems that IBM is living with, with Oracle's living with, and others. He's running on Oracle, right? I mean, he's running on Oracle. Exactly. Well, yeah, parts, I mean, he's running at Rails and so he's running some open-source components, but you're correct in saying that. And so, you know, and that's the thing is when you're a publicly-traded company, it's easy to criticize, but when you're running a publicly-traded company, it's very hard for you to go to your investors and say, guys, would you give me a break for a year or two years and understanding when I'm revamping some of this and I'm changing some of that? And the profitability may not be there while I do that. Wall Street doesn't want to hear that. So since we're on a roll here, what do you think of these companies? I got to ask you about Workday. I mean, this is a company that has seemingly executed every single step of the way, I mean, surpassing many expectations. What's your take on Workday? So I'm a big fan and again, kind of ironic, I actually got to know Dave. Dave used to work for you, no. Right? No, I wish. He refused to work for you. But Dave actually, when I was at IBM, Dave was a customer. He had a company with an MRP package that ran on the AS400. And I was the AS400 tools developer, the manager for that in the Toronto lab. And so we met and he said, Henry, will this ever be client server? You gotta remember, this is a 1988 conversation. Yeah, right. And he was across the bay here and Walnut Creek to be precise. And I met with him and his brother and I said, Dave, I may get you from a green screen to a color screen, but that's about it. It's a dumb head, no functionality, no distribution, all the functionality is on the AS400. And he said, thank you very much. And didn't hear from him. And one day I read the papers and he sold the company. And because he couldn't migrate his install base to where he thought the world was going, which is client server, moved up the street, started PeopleSoft. And he wrote that one. And one would argue that says, Larry thinks he won, but I don't really know if Larry won when he bought PeopleSoft. Because Dave then took that and had no install base to worry about. Plenty of capital. Gave him a time bomb. Best thing that happened to him, right? And went and started Workday. And Workday was engineered with all the latest technology. It's an object-oriented database. I mean, he's really, and I can tell you personally, I have a friend of mine that's, and I won't name the company, but pretty reasonable IT budget. It's a $2.3 billion budget annual. And he put in Workday. And I said, so any issues? He says, yeah, I've got an issue. He said, the end users love it so much that they're forcing me to replace PeopleSoft at a faster rate than I was ready to replace it in other geographies. So to answer your question, just look at the number. Right to me, the biggest test is, are the customers buying it? How do they feel about it? And so they've done a very, very good job of understanding exactly what the end user wanted and built the right application. Let's talk about the consumerization trend that's going on. You mentioned that, you know, it brings me, you talk about the dumb note at the end, the head of the network. Now it's getting more functionality. The consumerization trend, consumerization of IT, mobile devices. You see Google out there. What do you think of like Googles of the world who are saying, hey, you know, we're just going to go to the consumer layer and that soon will be adopted by the enterprise. And how does that fit into the cloud and the trends we're seeing with Amazon, for instance. So Amazon and Google, what's your comments on those? Yes, I'm not a real expert on the consumer side. I'm more of an enterprise guy, but if you look at what they've done, if you look at what Google is, I mean, if you look at the amount, most people don't realize they own their own fiber from here to Asia Pacific, from here across the continent, across the continent into Europe. Really what they've done is if you really look at it, if you want to paint it this way, you can put a cover, a metal cover around the world and really Google's inside as a computer. Where does a computer? It just distributed the functionality across the world. I think I have a lot of respect for Google and I think that, you know, some people are scared of them, but I can tell you that frankly, I'm amazed at what they've done. And of course they're always pushing the envelope in regards to moving into in memory, right? And moving and pushing the envelope on the technology. So I have a lot of respect for them. I mean, I'm sure your user Google, your user Google. It's interesting to see what they're doing with the database, not only what they do with Bigtable, but you're seeing, you know, new models of database that they're putting forth with things like Spanner. And, you know, it's fascinating. Now you've also helped incubate, invest in, advise a number of companies, Pentaho, Revolution Analytics, Enos and others. And you mentioned, we were talking about open source. You've got a background in open source, even though you came out of the mainframe world. So I wonder if you could comment on the whole open source movement, the new sort of trend that that's bringing to the world and where that fits into your sort of investment strategy. Okay, so it was kind of coincidentally, I was asked by one of the board members of via Linux to sit on the board. And so, and I got to know the A and VA, Larry Augustin. He's the A and VA. And so being on the board, I was always intrigued by what source forage. Hey, geez, look at these projects. And then I started to go through these projects. And I said, you know, if you stitch some of these projects together, we could probably build a platform and compete with some of my ex-employers. And in fact, not to get too distracted, I was on the NT advisory board in the early 90s for Microsoft and Gates used to tell us all the time, you don't capitalize your market, I will. And it always stuck in my mind. I said, well, you know, why can't we use open source to little capitalize and lower the cost? So that's when we formed Pentaho. So we were a bunch of ex-cognos, Hyperion, and so on founders. And so we built this free software, you cobble together, right? It's free, but it's not that easy because first you have to learn this in 2004 when we founded the company is, you know, when we went to the law firm said we want to buy these open source projects. So we want to buy Kettle out of Belgium. We want to buy a JFree reports out of Germany. We want to buy a no-lap engine here in town and say, well, Mondrian, how do we do that? What am I buying? Is it copyrights? Is it license? Lawyers didn't even know at the time. So we had to start working with the lawyers on it. You were president of creation. Yeah. And to write the contracts. And we acquired those properties and we did it. And the other thing that was interesting, and again, a lot of it's luck in this business, I happened to live in Raleigh at the time. And of course, I'm Canadian and the founder of Red Hat's Canadian and I got to see how Red Hat did it. And so you figured it out and I asked him, I said, I don't understand why the U.S. government is your largest customer at that time. I mean, it's open source. And he said, exactly. I said, no, no, I, well, okay, I don't get the exactly. And he said, well, in a proprietary operating system, when let's say, use Windows and I don't want to, you know, Windows or Solaris, I'm not picking on any of them, but you know, let's say somebody decides a hacker decides to go and hack into that system. It takes a lot of complaints before they start saying, geez, you know, I think maybe somebody did break into here. Let's look into this. And the last thing they want to do is put a public message out there. Somebody broke into the operating system and there's a malicious code in there and it could be hacking. In the open source world, here it is. I mean, I learned that at Pentado. We did a nightly build, you put it out there. And while you're sleeping, there's always an engineer who wants to show they're smarter than your engineer. And so, let's pick an example. Somebody in New Zealand is taking your code apart. It's open source. And then they call you and they say, you know what, they don't even call you. By the time you get up, there's a message out there and they say, hey, you know what, I found this problem. And if you handle it right, you say, good catch, would you contribute that fix? And I want to incorporate it. And so, you're constantly getting people looking at your code. So, getting back to this analogy with Red Hat, they're saying, who in their right mind is going to go put a malicious bug in Linux when you've got thousands of people every night going through this code? They're going to find them immediately. And by the way, the open source community does cleanse itself. And taboo, you know, just get rid of this, shun this individual. It washes it out of the system. So, I never even thought of that aspect. So, open source, you can get contributions. You're constantly having people doing QA on it, right? And from a security perspective, it's open. So, it's secure. So, there's a lot of advantage. And so, I started to, you know, so obviously Pentaho was, you know, reporting an analysis at the time. Xenos was a systems management. Compierre was ERP. So, I found that certain models work, and certain industries work, and some don't. And so, again, if you just follow us. He'll fix that for you. So, Dave, I want to get to Dave. So, Dave, what's your take on his comments about work, Dave? That makes total sense to what they said. Well, so, you're seeing very interesting sort of trends here in the Valley, obviously. I mean, we had Actian on earlier, and I'm interested, Andre, in sort of what led you to Actian. Essentially, you know, spending, let's say, upwards of $100 million rolling up, you know, various parts of the company. We were talking, John, you know, it sounds like a lot of money. It is a lot of money. But actually, it's really not. If you can make clever investments, you can get to market much, much, much sooner. So, what did lead you to Actian? How did you get there? So, actually, Terry Garnett. So, Terry and I got together, and Terry said, look, I bought this property called Ingress. And so, I've got a very good relational database that's generating good cash, right? Because, you know, people don't append in databases that work, right? So, there was a good install base. And then, we determined that says, look, the relational model has probably been pushed as far as it can. Oracle may not agree with that, but, you know, we do think that the relational model has been pushed to areas where it's not really suitable. So, then, we bought another company out of Germany on the object-oriented database. And we said, okay, for certain applications, an object-oriented database is the right way to go. Then, we bought VectorWise out of Holland, which is really an SMP, columnar database for analytics. And then, we bought Park Cell, which is massively parallel, okay? Database, right, for analytics. And by the way, it's the foundation, as you know, of Redshift, which is what Amazon is using. So, we have people today that develop and test on Amazon and deploy in premise. And some do deploy also on Amazon as well. So, they do everything on Amazon. So, it's not a roll-up in the sense of a CA where you build multiple products of the same nature and you just economy of scales. They were taught through acquisitions of how do you put the pieces together. And pervasive was the crown jewel, where you said, okay, you got to be able to integrate data, right, both on premise and in the cloud. And that was the piece that we got from pervasive. And John called it data fusion. Exactly. So, take us through the dialogue. I mean, obviously, you see the opportunity, you mentioned in your story about the open source, how you see the opportunity to get these open source projects same and acting and the market's shifting. You guys saw an opportunity to pull all this together. I mean, was it a fast conversation? Was it pretty much? Here's what we want to do. Was there a lot of strategic planning involved? No, the advantage, you know, Terry's a pretty smart guy. He gets it pretty quickly and Terry's a fabulous to work with. And so, no, these decisions got made pretty quickly. And that's the advantage of being private. We are at a size and profit where if we wanted to go public, we could. And we're not yet. And the reason we're not yet is we still have some work that we want to do. And when you're private, there's lots of things you can do that you don't have to share outside the family. And we're not finished yet. So when big data came along, by the way, so let's get to the firm big data and maybe tie it back to the beginning comments I made, what you're going to see now is an advent of a new position in many corporations like CDO, Chief Data Officer. And in fact, I was with the Chief Data Officer of Barclay's yesterday here in town. And the reason- He was on theCUBE. Was that on theCUBE? Was it? Osama Fayad is the CDO. Osama Fayad. Okay, anyway, we'll come back to it. Sorry. Something was up. But, and the reason that's happening- CD Ameritrade, I'm sorry. No, that's okay. So, but what's happening though- Barclay couldn't make it because of that library issue. That's the word. That's the word. They have their challenges. But so what's happening though, as I said, is I live through data processing managers to then CIOs where their job was to extract value out of the information initially in one silo then across multiple silos, but always within an enterprise. And now the CDO's job is to say, okay, but I've got to be able to tap the big data that's coming in, okay, from other sources, okay? So, and I mean, and as you well know, it's everything from road sensors. I mean, I could go on and on about different types of applications that I've seen. And so, ADUPE is a, to us, is just another great data store where you first put the information in. And then in some cases, you'll be able to do some processing within ADUPE, but we also believe that in some cases, you'll want to extract that and put it into, like an Action platform than to do some- So I love this conversation about the CDO. We do a conference every year with the folks at MIT on the Chief Data Officer. And it's something that we agree with. We believe that that role is emerging and will emerge. It's not mainstream yet. My question is, do you think the CDO, the Chief Data Officer, should report to the CIO or is there an independent position? Why or why not? So it's funny, I'm working with a large pharmaceutical company right now and helping them define that. And the way they're going to do it is it's actually going to be parallel. So the CIO is really going to run the internal and the CDO is actually going to work very close to the CIO but in parallel. And because their mission is somewhat different. And so in this case, by the way, and I'll leave it again nameless, but it's a large pharmaceutical company that decided that they no longer want to spend a lot of money marketing, or I guess it's marketing, marketing, not selling, marketing too, and informing doctors about why they should prescribe their drugs. They've decided to do is they took a billion dollars, which sounds like a lot of money, but it's small in their world, a billion dollars out of their sales and marketing budget. And the CEO has allocated that now to a new project. They want to hire a CDO and the CDO has got to figure out how do I get directly to the end user, the potential end user of my product and influence that decision. So I don't know if you've noticed lately but you'll see more ads on TV. Pharmacists are actually advertising more directly on TV. That's turned out to be actually quite good. So they're now going back to the doctors. I saw this ad. Doctor, why would you not recommend this one? So now with that, they're starting to track who these people are, they want to track but who did read that ad? And they wanted to have a dialogue and basically develop a relationship directly with the potential end user patient of that drug. So I got just a plug. So our friend, Dr. Professor Richard Wang is doing a big study right now on Chief Data Officer. So if you're a Chief Data Officer, contact him and he wants to hear from you. I'd be more than happy to share that. And that's one example. So you're talking about big data. So you can imagine, they want to do sentiment analysis, what's being said about their drugs and the marketplace. It's also got legal implication, okay? Are they hinting that there may be a problem with the drug that we've already got out there that we should be jumping on, right? So proactively before the FDA does. By the way, what's our competition saying about our drugs? What are the patients saying about it? What are the trends and so on? So you can imagine the amount of things you've got to monitor in a company of that size. Andre, I want to ask you, what are you working on now? I mean, we're doing any investing? Are you on more boards? What's the key things that keeping you busy these days? Well, I'm on eight different boards, mainly on the technology. One of them is publicly traded. So I'm chairman of a publicly traded company and the rest are all private. Are you doing any investments at all? Yeah, I co-invest in certain deals. And so, but big data to me. So when people said, well, Andre, big data is changing the world. I said, guys, let's not, it is, but what's really changed is the fact that most people are starting to realize, I bet you we can walk outside this hotel. Let's go to the street. I'll show you some cut marks all over the pavement where they've embedded sensors in the road. I bet you that's only happened a few years ago. Figure out every sensor that there is, okay? I was talking to a, and I'll leave it nameless at this point, but a tire manufacturer that wants to actually, around a tire, put a copper bead, okay? Five across, five deep. So as you're driving your car, it can figure out if it's wearing out the outer bead or the inner bead, the car's not aligned properly. I'm gonna send you a text message that, you know, your car's not aligned properly, you should bring it in. And as you start wearing down the depthness, I can tell, I said, by based on your driving habit, and I know what your mileage is, driving your, you are gonna be out of compliance. In other words, your car's not gonna pass inspection at the next inspection, and it's not safe, bring it in and we'll change the tires for you. That's gonna help the insurance companies, too. Look at those driving patterns, too. So figure out, that's four tires per car. How many sensor points? That's just the tires, right? So figure out the oil, the usage of oil, and so on. We can go on and on, but just, so one application I was looking at in Detroit, the guys laugh at petabytes as it, petabytes is a rounding error, okay? And so you figure out how many of those, the cars and the intelligence that are being built. So when you, you talk about big data, so to get to the point of big data, there's a lot of exciting technologies that are being funded around on how do you extrapolate the value out of big data, right? And there's gonna be some of the players like Acti on in the game, there's obviously gonna be some new technologies. New technology companies that we may decide to partner with or we may decide to acquire, right? But we're obviously keeping the eyes on and like everybody else of who we're gonna win and who we're gonna lose and which ones are gonna make. Quarter of the day, John. Petabytes is a rounding error. Yeah. Andre, we had a great conversation with Acti and going back to big data NYC and you guys, really impressive. Love the combination, love how you guys put it together. And again, you get a lot of room to grow, continue to do some great things. Thanks for coming on theCUBE. Thanks for sharing your amazing stories. It was awesome. We made the time go by really fast. Appreciate it. Well, great. Thank you very much, John. I appreciate it. Okay, this is theCUBE. We'll be right back with our next guest after this short break. I'm John Furrier with Dave Vellante, live in Silicon Valley for big data. S.V. Blue, right back.