 We're back live here towards the end of day one of Hadoop Summit 2012. This is siliconangle.com, siliconangle.tv, and you can judge with wikibon.org. This is theCUBE, our flagship telecast. We go out to the event, talk to everyone we can find, find the stories, find the knowledge, share that with you, expecting the signal from the noise, and we're happy to be here. I'm John Furrier, the founder of SiliconANGLE. I'm joined by Jeff Kelly, with wikibon.org. Jeff, we're back, end of the day. We are, we still got some time to go. We're squeezing in a couple more interviews, so what do we have here? Well, actually, I'm pretty excited about this interview. We've got Jim Kobielos from IBM, former Forrester research analyst, covering big data, now working at IBM doing some marketing around big data in the TISA, and your whole big data kind of strategy there. I'm an evangelist. You're an evangelist, and the reason I'm excited about this is because in my past life at TechTarget as a reporter, you are one of my greatest resources, and I really appreciate all of you. And you're one of my greatest reporters. Thank you very much for scratching my back there. It's a love fest here on theCUBE, but anyway, very excited. Tell us a little bit about what you're doing now at IBM. Yeah, at IBM, I'm a big data evangelist, and marketing evangelization, really, it's all about calling out what IBM, well, first of all, calling out mistakes for big data. How it's big data and big data analytics, advanced analytics, things like Hadoop, are being incorporated into business, modern business, to help you run your business better. So really calling out the applications, the use cases for big data. Of course, calling out what IBM has to offer in terms of products and services all around big data. And half of my position is marketing, blogging, tweeting, doing tweet chats, thought leadership papers, speaking. I'll be speaking here at the Hadoop Summit tomorrow. And about half of it is strategy. I work within and among the product teams. We have a lot of people doing marketing and strategy at IBM. I'm another voice around the virtual table, trying to nudge us in the right direction to deliver ever deeper solution portfolios for all things big data. So that's what I do. So tell us about. Well, you're very ranked high in our community. I see you on Twitter, but in our little Twitter tool we have, you're very prolific in big data, which is good for IBM is known internally. A lot of people blogging, a lot of social media within IBM, so. Big data. I've got a big mouth. I got big tweeting fingers here too. You can't shut me up. So tell us about, all right, so IBM here, we're here at Hortonworks Hadoop Summit. So we've got this kind of this upstart Hadoop community. We've got Hortonworks, Cladera and the likes. Then we've got IBM with a really vast and interesting portfolio of big data tools and services. So how does IBM view these guys like Hortonworks, like Cladera, who are kind of developing a focus exclusively kind of on the core of Hadoop. And how does that fit in IBM's big data strategy? Yeah, well we have a Hadoop offering, of course, big insights. So we've been in the Hadoop space for several years now with commercial solutions. And of course there are plenty of other solution providers out in the Hadoop and big data ecosystem that offer tools that either compete with us directly or compliment us and so forth. Really, you know, the Claderas and the Hortonworks and the MapRs and others, we're all part of the same community. We're all part of the community that has tools and products that address this new approach to doing advanced analytics in a massively parallel way, as it's called Hadoop. The market for Hadoop solutions continues to expand, continues to develop. The ecosystem vendors solution providers, service providers with different emphases, a lot of innovations going on. We're innovating, those guys are innovating. We just want to be part of this community. And we respect totally what the Claderas and the Hortonworks are doing. And it's an exciting time. So IBM, we have a lot to offer. And we make our case through all channels for why a customer should go with IBM. But IBM partners widely too. We have a lot of partners here at the show. We know the IBM challenges. We were just at the IBM Edge and there's IOD, there's innovation. IBM does not have a problem with marketing at the highest level. They are really, really strong on the brand marketing. But as you get into the nuances of the different verticals, there's a lot of crossover. So we're just at the Edge Storage Conference. It was nice to see that holistic view of a portfolio where the storage guys are talking about Tivoli and talking about big insight. So you're seeing that coalescing around IBM. Is that part of the new mindset of IBM? You have a new CEO. What's the internal vibe around IBM? Is it moving from the silo to more of a packaged groups? Because big data spans a couple different things. It's platformization, yeah. So the different platforms for big data. We have a really cool graphic that lays out at the very highest level. It'll be in my presentation tomorrow. The IBM Big Data Platform. It's a bit of Hadoop. They're big insights. It's a bit of data warehousing. We have Natisa. We have IBM Smart Analytics System. It's also stream computing. We have a product called InfoSphere Streams, which, oh, by the way, can execute MapReduce models in real time for complex event processing style applications. So really what IBM has is a range of Big Data Platforms to address the three V's and so forth. It has to do other vendors in various combinations, but we've got a fairly extensive set of platforms and tools to build MapReduce and then pagan so forth jobs to execute in various and sundry Big Data Platforms. So we offer a degree of choice for customers of the right platform for various big data applications. So in many ways, IBM has not only platforms, we have applications. We have SPSS. We have Cognos. We have a lot of things. We have appliances that can accept. You've got everything. IBM's got everything. We have professional services with business analytics. And also you guys have this massive customer scale. You have customers at such a high scale. I mean, we're here at Hadoop Summit where it's such a good feeling because it's like innovation, startups, it's an ecosystem that's growing, but they'd die to have one of IBM's littlest customers. Any of these startups. We've been around for a hundred plus years. So clearly we've got a huge history, lots of customers all over the world that have been building and scaling their databases, transactional analytics, year after year, everywhere in the world, the average size of databases continues to grow. And you've got Watson out there, great marketing showcase. Watson, yeah, Watson's more than marketing stunt on jeopardy. It's a real deal. It's an R&D program that has produced results that have been productized in solutions for particular vertical applications and things like healthcare. So Watson, that portfolio of intellectual property and R&D continues to grow. And we continue to productize it. So that's part of the overall solution value add that an IBM has to offer. Yeah, and I would share the audience out there. Go to YouTube. And Watson has Hadoop at its core, by the way. Yeah, I mean, so IBM, which is just amazing. YouTube.com slash SiliconANGLE. There's an interview on there with Jeff Jonas from IBM. We had in Orlando at the IBM Edge event. And that probably was at a storage event, which was, again, holistic, really was about big data. I mean, he's doing some stuff in IBM that is mind blowing big data. So I think there's a lot of people don't know that IBM's a big data company, have it wrong. They were really impressed. You can watch that video here. Have you noticed that all the storage vendors in the last several years have repositioned themselves as big data companies? Because where is the data going to be stored in massively parallel petabyte, exabyte? Flash-based. Yattabyte, whatever. Yatt-intel cores. I mean, if you look at the, you did what Intel's doing with all their multiple cores and you got flash, the storage architecture configuration or the assembly of storage architecture is changing. Multiple caches, cache this, cache everywhere, flash memory. So we're seeing that in memory persistence. And we heard it this morning on theCUBE about that from a developer standpoint. The notion that I can put petabytes in, quote, memory. You can't, yeah. You can't yet anywhere put, persists a petabyte of data in cache in a completely unified way. I think the industry is moving towards that. We were kind of fantasizing, but it ain't there yet. I mean, yeah, so yeah. Not there yet, but. We can't do it, nobody else can do it yet, yet. Not yet, but that's what's coming around the corner. Gigs and terabytes, it's impressive. So with that, how do you guys talk to the communities here? Because obviously it's great to be involved. We had Val from NetApp on and there's real good alignment between the big companies and these growing ecosystems like Hadoop. There doesn't seem to be any static at all. All the storage vendors like NetApp and so forth, they are very much trying to grow their stack in terms of analytic databases and modeling tools. So a number of them are partnering with the Hadoop providers of the world to build ever more scalable and high performance full solution stacks. And of course we've done that as well. So clearly we have within our own product portfolio all the piece parts to build a complete solution stack that also has best of breed, low cost storage at its very heart. Like I said, all the storage vendors are trying to deepen through partnerships. IBM's got all these products in the house. And of course we're doing deep R&D to continue to evolve it. One of the things, in addition to futuristic stuff like petabyte, the coherent cache data, multi-server for big data, for real-time big data, then that'll come in a few years time also on the storage side. I think one of the frontiers in big data will be storing full motion video and doing analytics against it in real time. We're talking beyond petabytes there. We're talking about exabytes and Yata bytes and so forth. That's something that by the end of this decade, we're going to see, I call it, this is just me, I call it big media. There's big data and data is data. It's structured and unstructured, but media, video and audio and all that, that's going to need to be stored and analyzed in real time. In a world where all media is converging digitally and where analytics in line to video distribution will be critically important, not just in the media industries, but in every industry that relies on streaming media to do training and to get its message out. So that's one of those futuristic things that we at IBM are looking into, the whole notion that full motion video as an object, type of object, will need to be brought into a broader big data platform story. That's just one IBM evangelist talking about what needs to happen. I'm not making any claims for what we might be doing. Well you don't officially speak for the company that's always the disclaimer, but no one ever got fired. I'm just winging it here, I'm just freelancing. No one ever got fired for being in the cube sharing their knowledge of the world, as they say. So thank you. They pay me to be a space cadet. Thanks for coming on the cube, we appreciate it. Sharing that knowledge with you, good analyst perspective, good social media mojo. Great, we'll see you on Twitter. Very prolific and big data at IBM, big player. We'll be right back with our next guest right after the short break.