 Okay, everybody. We're back. We're live in Santa Clara, California strata Day one. This is our second year at the strata conference and I'm here with Alex Williams and Jeff Kelly Alex of course is the editor-in-chief of services angle and Jeff is wiki bonds Lead big data analyst gentlemen. Welcome back. Thanks for hanging with us here. It's It's nine o'clock East Coast time. So thank you for hanging with us as well and We've covered a little bit of ground today We did a lot of sort of setting up for the the day we talked about the the marketplace and Alex We just had a really interesting guest on Ron former quad cast now would think big analytics talking about some of the important services aspects of Big data and Jeff in your report services was I think you know 40 45 percent of the pie. Yeah, so Services always the biggest business, isn't it? It's gonna play a huge role in big data because as we all know there's a huge talent shortage Both that kind of the Hadoop the engineering level as well as data science So service is gonna play a huge role both on a technical Technical areas as well as I'm actually doing some of the analytics talk a little bit about that in what Shape and form are you and Alex you seeing services and how is it different from the kind of traditional break-fix stuff? well you know, I was just out talking to some folks from impetus and they're a consulting group and they focus on big data and doing analytics and they've been doing it for a few years and What's interesting is like they have this big, you know, they O'Reilly actually has You know a something you can put on your lanyard that says we're hiring and every one of them had it Oh, I need one of those. Yeah, and and I want I walked out. I say you don't want to hire me but listen I got some questions for you and What what they're saying is that there is just just a huge huge shortage and I think what it goes back to what we were hearing Said earlier with some of the guests that education is just critical and There's a there's a lacking ecosystem out there And those two things combined make it Very difficult for companies to adopt big data really at all and so we're you know It's may sound cliche, but yes, we are right at the beginning here Yeah, I mean I would agree with that. I think I mean one of the one of the big one of the things holding big data back is a lack of understanding just confusion among traditional enterprise what is big data? How is it going to fit into my? current Data infrastructure because you know these big companies have invested millions and millions of dollars in data warehousing bi technology And they don't want to just throw that out So you know there's a lot of confusion about how big data is going to fit into your current landscape You know what use cases are big or big data use cases what use cases are maybe more traditional data warehousing use cases so yeah, I mean there's just a Lack of education around big data and that that needs to needs to education needs to happen I think there's a there's a very much of a large cultural shift though that will Make this much more accessible and I think you see it in such things as and May seem like Abstract but like the maker movement and the maker movement is all about bringing the software into like Tables and chairs and and you know in vehicles and doing all these different kinds of things They really require very sophisticated technologies But also using lots and lots of data and so really this culture of Doing which is really about the web this culture of doing and of creating things That's going to really going to be the imbattis it and and it'll have to grow organically So we have a number of guests tomorrow. We're going to kick it off with Mike Olson of cloud era. He's the CEO Mike Olson was one of the first people that I interviewed with my co-host John Furrier down at Hadoop world in 2000 and when was that 2010 and Mike gave me an education on Hadoop and I said wow after being at Hadoop world This is something that we really have to invest in at Wikibon That's really how we brought on Jeff Kelly and and you know, it's exploded since then with Wikibon and silicon angle So Mike's coming on and we're going to have Amy O'Connor of Nokia who was on at Hadoop world in Last fall in New York City a practitioner of Hadoop and She's coming on Chris Moody from Gnip Gnip sells data like the Twitter firehose So excited about having those guys on some of the other folks Data stacks is coming on Sandy Steyer of 1010 data Rob Metcalf not the Bob Metcalf a different Bob Metcalf of digital reasoning Pentaho is going to be on talent Shmarzo's coming back for another appearance Jack Norris VP of marketing at map bar We got some really interesting discussions. We've had with him and excited to have him on Talking about that whole ecosystem development at what's happening there revolution analytics Dave rich who's the CEO is coming on James Phillips who's been on before couch base. They're doing some interesting things and and a number of other guests VCs will be on some some drive-bys no doubt people that we we see we've got some room in the schedule And then spotlights Jeff we got some couple of spotlights coming up this week talk about those a little bit Right well one you mentioned cladera Is a spotlight sponsor the cube so we're gonna we're gonna speaking with Mike as we mentioned as you mentioned We're gonna talk a little bit about some of the news that cladera has made in the last couple of days But it's gonna move into kind of gonna explore the the real-world use cases for Hadoop at this point They're really being you how they're really being used so spotlights are topical and as Jeff said they're sponsored So we're gonna talk about real-world use cases. What's what's the other spotlight? Who's sponsoring it? What's the other spotlight we're also talking to a digital reasoning now They do some very interesting work their product their platform called synthesis essentially Automates the process of understanding human communication whether it be emails texts whatever it may be they work with a lot of three letter agencies Intelligence agencies with the government which I can't tell you what they are I'd have to kill you But they do some really interesting stuff and they're looking they're starting to expand into the enterprise So I think we're gonna talk a little bit about how that technology can be applied To more to two enterprise use cases beyond just kind of what we're here seeing in the intelligence community Hey, John Furrier was talking earlier today. He said that you know, we had talked about 2010 was the year of what is Hadoop? 2011 is you know big data is sort of the new economy big data monetization is sort of the future of our business and John said that 2012 is good about two things platform maturity and application innovation and and so I Would like to hear you guys hear your opinion of that. What what does that mean? So so I think he's right, but but let's peel the onion skin on that a little bit Let's start with with platform Maturity, what are you guys thinking about platform maturity? Where are we at with platform maturity and and and why is 2012 gonna be special if Furrier's prediction is correct Platform maturity in terms of platform as a service, you know providers we're seeing a you know just a blossoming of that of that part of the market and I think so far. We're seeing leaders like cloud foundry really taking taking a charge Heroku is still an established player there and You know, you have companies like engine yard Plus you have others like Tier tier 3 and these companies I think really do represent the feet Represent the future though their their revenues are really quite small, but what they're doing essentially is there They're abstracting a lot of the complexity that comes with developing applications and so there's that real connection there and when those But the problem is and I think it's still when we see a lot is this this misunderstanding about how platforms actually relate to the way that you would you know in in in more traditional senses, you know build your own stack and so There's this issue really what it comes down to is latency and latency is is really becoming the big issue because You may think, you know, oh, yeah As you know as the platform for oh, yeah It's so much easier to you know to use our you know use our platform But if it's if there's any kind of latency people relating that to like Any website when it's slow, you know, so they think it's not really working that well and they don't really see the contextual difference between provisioning their own stack and seeing another website how per compares to the platform so That's gonna be I think there's gonna be some need to really mature that platform technology So the application development doesn't seem like a laborious process So Jeff in your big data predictions post last year you said that applications We're gonna be where all the action is application innovation Doesn't the platform have to mature before that can happen or can they happen in parallel? Well, I think they can happen in parallel You know, I think Hadoop is a mature enough Platform at this point that you are you're already seeing a lot of big data startup big data application startups start at the market developing Applications that leverage all this unstructured data that we're seeing proliferate on the web and mobile devices sensors, etc you know in terms of Bringing Hadoop into a more traditional enterprise environment. I think that's that's one area where this this year is very important We're we're gonna see a lot of proof of concept Hadoop deployments Go into production this year or try to go into production this year So dealing with all the scale scalability issues the security issues user access things like that uptime requirements That is where Hadoop needs to prove itself and I think that's where we're gonna see this year It's kind of kind of gonna be a make-or-break in a lot of ways for Hadoop as a platform in an enterprise environment as a stable and reliable platform Okay, and then of course the other thing is that John mentioned is that 2013 would be the year of monetization I don't know if he said it that way, but the year of making money Now Jeff if you look at your s-curve in you know your ogive curve ogive curve in the In the wiki bond report on big data. Let me just pull it up here big data revenue and And it's everywhere this this report is all over the place So big data is big business so When you look at that that s-curve 2013 Furrier's calling for the year of really you know making big bucks Do you buy that is it gonna take a little longer? I think it could take a little bit longer, but So here's your curve right the so based on this the steep part of the s really comes in a little later 2014 2015 so Furrier's calling it 2013. I guess in beyond What's your take on that? You know I think You know we are gonna see a significant uptick and as you as you can see in our our report But I think it is gonna take a little bit longer in terms of really monetizing Big data because as I said, this is the year where we're really so gonna start to see Use cases or I should say deployments that are at scale and production and and that is where we need to prove ourselves once that is proven and we get to a point where Hadoop is not a unknown when it comes to Enterprise readiness then we're gonna see I believe a pretty steep steep climb as you see there in our in our report in terms of vendors actually Making a significant amount of money on this technology. The other thing of course is the you know the application Development space. I mean, there's just so many opportunities One of the you know the most exciting thing to me about this market is so many opportunities I mean sky's the limit in terms of what you can do with this data You know not all these big data applications startups are gonna are gonna succeed and certainly Many of them won't but some will and there's gonna be some really interesting applications coming out over the next couple years As those start to mature You're really gonna start to see them The ones that do kind of succeed the one that find really impactful business problems they can solve You're gonna see them making some significant money in this market. So Alex you wrote a post her guess earlier today So that's some of the demographics and the stats of people traveling to the event here at strata You said developers travel an average of two thousand three hundred forty six miles to attend the conference That's something I was surprised. I'm too. I think it shows kind of the Worldwide community that you have here. I think you probably have people coming from all corners of the world here 83% of the attendees are male. It's not so unusual, but it's typical tech conference 33% of the attendees said the organization stores an average of one terabyte of data per day Produced each. It doesn't say day. Yeah to update that each day. Yeah each terabyte of data per day Yeah, yeah now. That's more like it. Yeah, cuz we produce how much data we produce today mark Hey mark Hopkins, how much data we produce today and an average day of the cube a Quarter terabyte a day and that's today Today was like half day half day of broadcasting is a quarter of a terabyte so full day half terabyte right and That's pretty good. That's pretty significant Yeah So this is yeah, this is the data conference, you know, you can tell by those numbers Yeah, good. What else are you guys working on over at services angle and DevOps angle at services angle Working on a few stories and doing more profiles on the storage players. I think the storage is the is an interesting You know, it's just an interesting story overall There's a lot of interesting companies out there and it relates so much to the the competitive landscape Writing about Hadoop and big data writing more writing a longer post on Hadoop services We interviewed the emcee a data scientist and and we're starting looking to look for some customers here in particular Plus we're also, you know, continue to work on, you know, those stories that are about, you know, that what's happening With these leading edge cloud services providers or even these service providers that offer or software providers offer for instance No SQL, you know databases and because that's really Interesting in terms of what's really catching on and seems like no SQL is really catching on do you think the The data lifecycle is changing as a result of all this this big data, you know Kind of used to be okay I I use it and and you know, maybe I Maybe I use it over, you know, every every month or every quarter and I use the data But but and then I archive it or you know back it up archive it. Is that data cycle? Changing because of the ability to ingest all this information into something like Hadoop or is there a similar paradigm? I'll just say something quickly than Jeff can add it I mean definitely the tearing of data is really happening We're really seeing that and with that tearing of data. We're seeing different, you know methods for networking Network virtualization, you know, so I definitely think that data is much more present in the daily conversation and we're really figuring out ways for instance to use technologies like Hadoop To be a backup, you know environment for for data itself So it it just I think it really is it kind of a consuming topic for People on a much more frequent basis Yeah, I would agree I mean One of the beauties of Hadoop is it allows you to store pretty much all your data and make it available to access for Historical analysis When it will whenever you want as opposed to kind of archiving your data and it kind of sits there and isn't used anymore So in that sense Absolutely it changes the life cycle and it also changes with the other thing we need to change However, of course as our friend Abby met it from Triseta pointed out to me You know people have this mindset that you know We only have a Limited mindset in terms of what is available for analysis and it's gonna we need to break through and and continue the education To make people understand what to do. Well, you can store everything if you want to theoretically so absolutely it's gonna change The data life cycle, but that has to be coupled with people understanding that change Understanding all the new possibilities And understanding that now, you know, you don't have to sample data You can take all your historical data And run through it and in myriad ways in the words of Bobby meta. It's game-changing man Is he here? He is I haven't spoken to him yet, but he will be here for sure. He's coming on the queue He's he doesn't he doesn't have a definite slight yet, but he is definitely coming on He's committed to come on the queue. He's committed to come on the queue. He's keynoting tomorrow morning So I'm gonna see if I can round him up. I think also talk to him. He says yes, I'll be I'll come on I'll get some time and come on. Oh, yeah, absolutely. All right, so it wouldn't be a big day to come I think there's there's also the opportunity this come this this conference to really find some of these hidden gems these Practitioners out there who are doing some really innovative stuff the people linked in the people who are at Bosho You know these companies that are really very very innovative and their practitioners are living this stuff every day Well Twitter's coming on right? We've got one of the lead engineers at Twitter coming on the Yeah, Nathan Mars, who's the who's the gentleman who actually authored storm, so that'll be Okay, that's good their real-time Analytics ended so that'll be very interesting. Yeah, I agree I mean that one of the things about the market a young market like this is those early adopters that are really pushing the envelope are By almost by definition doing innovative new things So anytime you can you can talk to some people that are actually doing doing this using this technology in the real World as such an early part early Area in terms of adoption You're gonna see some really interesting use cases some innovative approaches