 Live from Dublin, Ireland, it's theCUBE. Covering Hadoop Summit Europe 2016. Brought to you by Hortonworks. Now your host, Dave Vellante. Welcome back to Dublin, everybody. This is day two for us of Hadoop Summit 16 in Dublin. It's actually day four for the conference. First two days we're training. You know, a lot of the hardcore practitioners coming in and a lot of transfer of knowledge. And then yesterday was the big kickoff session. Keynotes today, more keynotes. Kerry James is here from EMC. Gave a keynote talk yesterday. Kerry, welcome. And also, we're joined by Mark Mason of Hortonworks. Thank you. Gentlemen, thanks for coming to theCUBE. Great to see you. Thanks for having us. Thanks for the opportunity. So this is great. I mean, this is my first Hadoop Summit in Europe. We've done a number, of course, in the States. Good crowd here, about 1,400 people. Good mix, a lot of practitioners, some business folks. Awesome venue, so. Shame about the weather. Mark, pretty good show. Yeah, the weather comes and it goes. It's Ireland, but Mark, the event last night was fantastic at the Guinness factory. I mean, who doesn't love that? From Hortonworks perspective, you got to be really pleased. I think we're incredibly pleased. Just to see the growth, as you said, and the attendees and the audience, I think the mix has been fantastic. And what we did actually in the first two days as well, we mentioned the other training, there was a big partner works part of the session as well. So the evolving and growing partner community in the ecosystem that we have together, they're all here. It's great seeing so many people, so many faces. Kerry, from your standpoint, we're seeing it do grow up, at least it's in adolescence. Right. And I remember, it was EMC World 2011, I want to say. And Pat Gelsinger was on, and I knew I guys had just done a deal with MapR before the whole pivotal strategy was all done. Pat said we're going to bring some adult supervision to the Hadoop world, we're going to bring the enterprise discipline. You were early on in that theme, many, many others sort of joined in, including the distribution benefit. We're doing that too. So it was a lot of R&D money and innovation that went in to bringing Hadoop to the enterprise. So talk about that a little bit and how the conversation has changed as a result. Sure, so I mean obviously for us when we first started the conversation, we're trying to understand how to bring enterprise values to Hadoop. At first it was really just around the storage mechanisms and how to drive those pieces. But as we've progressed over the last three and a half to four years in this journey for us as well, it's progressed beyond that, right? It's progressed just past the storage mechanisms to actually now come into providing enterprise into the data management pieces because getting the data in and storing it is one piece but it being useful to the data scientists where it's in a govern and a curated fashion, there's not just a large pool of information that you're trying to, as we say, try to muddle through the swamp, right? But bring that piece in there as well, as well as also bringing adult supervision to the ecosystem, to the compute side of the house as well, right? The traditional pieces you build out of 500,000 no clusters and you hit your cluster sprawl it became a huge administrative nightmare for IT which was, at the end of the day, it was kind of hampering some of the adoption of Hadoop. So also bringing some adult supervision to that side and helping to ease those pains, again, simplify the whole journey, has been good for us, right? And we've honestly, we've changed our position, as you said, the Hadoop distribution vendors have grown up pieces and we've now been able to leverage those partnerships where we can now complement versus having to kind of compete in those arenas. Well, it's interesting too, to see the whole evolution here. I mean, when you look at the original, you know, instantiations of Hadoop, it was really to drive costs down and do things that you couldn't do with traditional technology. I'm not going to buy a symmetric to take care of all those unstructured data. Right. And, you know, people that say, hey, it's a new storage mechanism. The thing I love about this business in general, but EMC in particular is, you know, you got a huge franchise. Right. But you got to innovate. You do have to innovate. And another part for us too, is that you said, right? Going back, it's kind of started with that storage and I'm not going to buy some metrics. I mean, for us too, the conversation from the business side has changed a lot too. Three, three and a half years ago, we were out evangelizing and trying to have them to convince people that there was value in doing these analytics, right? The IT guy saw the cost savings, but the business never saw the data as an asset. That, honestly, in the last nine months, I've not had that conversation. Now the conversation is evolving too again, is it's great, we understand there's value in it, but it's difficult for us to get there. So help us simplify those pieces so we can actually, you know, leapfrogs in the technical obstacles and actually get to the value that you and all the ecosystem continues to talk about. Yeah. And that's an interesting thing because like you said, a lot of the conversations early on about renovating, taking costs out of the business, and we can do that and we have some very interesting sales plays that we're working together and the unique values to take to market. But what we can see now is once we've realized that value, realized that cost saving, then we can actually use that money to then drive more transformational change and start driving new services or applications, drive more revenue into the business. So it's a really interesting shift in the go-to market. Is that predominantly what's happening and happening in the customer base? Is there a gain sharing? In other words, I got my, you know, old creaky enterprise data warehouse. I always sort of make jokes about that. But I'm dramatically lowering the cost of managing that data. Are companies just dropping that to the bottom line or as you're sort of implying, investing now on new opportunities? I just think- You guys can both address that. Well, I think that one of the key things that we're seeing is every single business is a data business. And the businesses that are going to do well are the organizations that embrace this and understand that using this data to drive them forward, create competitive advantage, they're the ones that are really going to start seeing the value early and people are going to get left behind, right? So I think this is something that's really transforming. Well, for decades, as you were saying before, Kerry, information was viewed as a liability. And, you know, the federal rules, the civil procedure, the states anyway, changed. And it became, oh, email, email, archiving, spend whatever I can. And it was the general council that was sort of running the show as it related to data and governance. And that flipped the last three or four or five years. Two or three years even. What are your thoughts on that? So, I mean, the thoughts around that, you're exactly right. So, I mean, early on, right, as you said, data was seen as a liability. I mean, think about back to the early days of COBOL, right? It was such a liability that you took the two-year digits out of the century codes just to save the money. Because managing it and storing it was so expensive. You know, with the advent, you know, the driving down, again, the storage side, you get this operational efficiencies. And that's great. That's one half the story. But again, as Mark was saying, right, to be able to actually take that. And some people are, on the IT side, seeing that purely as a cost savings. But what's interesting is you're seeing the business on the other side saying, I no longer see that as a cost savings to me, I now see that data as an asset. And I think that's where the model has changed, right? Which is where businesses, and a lot of times the business and marketing sides are actually having the bigger IT spins. And there's now seeing this data as an asset to their organization. How they can manage their products, how to better serve their customers. And, you know, a great example in the open market place is Uber, right? So Uber uses data to be a $50 billion company, and they own not a single event other than an application in data sets. So that's where that shift in seeing data as an asset and as a value generator is what's driven those changes in governance. And there's also an interesting shift, talking about shifts in the types of data that are available to an organization. And people have relied for the longest time on the more traditional data types in order to build a vision of what they're doing as a business, the customers they're serving. But now we have the opportunity to get data from multiple edge devices. We made a recent acquisition and now we have data at rest with Hadoop. But we also manage data in motion with their hallmarks data flow. So we're bringing all data under management. We can now enrich data in one place with multiple other data sets to try and, you know, transition the sort of things that we're looking at, build new insights. And like I said before about the applications and the services we can take to market, it's radically shifting. So let's turn that into sort of the business impact and maybe even some examples. So as I was saying the EDW, we've improved upon that. And you've seen many examples, certainly fraud detection, right? There's areas for improvement, but and it seems to be getting better even in the past 12 months. Fewer false positives, right? Sampling, the duple out sampling to, you know, we've killed sampling. Now we work on a whole dataset. I guess ad tech is a piece of that sort of real time, you know, nature, like it or hate it. Some people making money at it, that's great. Where are we seeing or will we see the business impacts and maybe you could give us some examples beyond just sort of making the data warehouse do things that you couldn't do before and do things faster? Where is that innovation from the end customer standpoint coming in? Examples, use cases? Yeah, so I mean, obviously a lot of that is being driven from the perspective of creating applications. So back to the Uber example, right? It's a great example out there. It's an application that allows you to actually interact in real time with that data. Geolocation information as well as, you know, information about the passenger, information about the driver. So all that's delivered to you in a real time. Again, back to Mark's comment about not just, you know, taking that data but also, you know, pushing it back to historical, right? So there was a little bit of a hiccup last year in one of the marketing things, right? Which is, well, I won't say who it was, but they're like, we know where you were on Valentine's Day. Wait a minute, because they're storing that historical data. That's a good example of that shift. Another shift for us we've seen is especially like the airline reservations, especially with the advent of all the different websites you can book information from, you know, you have to be able to manage and control that data in real time, because if I book a ticket, Mark books that ticket, he doesn't want to be sold out if it's really sold out before he books it. So be able to manage those pieces in real time and interact with that data from multiple different positions, multiple different venues, web, mobile, et cetera, is where we also see a huge shift in the value moving. There's an interesting use case that we talk about sometimes with an insurance company that we have in the US, and we'll find this spread across the globe, I think, as well, where an insurance company can have a device in your car, they can understand how you drive your car, the speeds that you drive at, how quickly you break, and they can start to understand your driving behavior and your patterns, which may or may not be a good thing, but what they can do then is create a bespoke insurance policy for you, a tailored premium for you, based on driving, and they can make certain savings, or you could make savings, or maybe not, if you're a safe driver, right? So you've got all of these different things where you understand different things about your customer, and you can drive new outcomes. Actually, I have a really good driving record, but I don't always drive slowly. But that's, when that first came out, I was my reaction, oh, crap, I don't want this device in my car, but the data that you gave yesterday was that that insurance company, and I think I know who it is, saved a fair amount of dough, I mean, it was billions, in terms of increased premiums and cut costs, and so, good examples. So, the tech that's enabling that, going from data lake, static data, the data in motion, and then other tooling on top of that, as well as applications, you know, I want to get your take on this. We always talk about the lack of applications on top of the due, but then we just, I wrote down Uber, Airbnb, LinkedIn, Spotify, these are all big data apps. We just don't think of them as big data apps. I don't know, maybe because they're not payroll, ERP, but that's not what we're trying to do. We're not trying to pave the cow path here. I wonder if you could talk about your perspective on what you're seeing in the ecosystem, in terms of innovation around whatever, the apps, the definition of an app. Well, I think, going back to a previous point, that every business is a data business, and I think using the data that you have in order to be relevant, be innovative, create new services and applications, I think that's a real driver for every single type of business. Uber, you mentioned Airbnb, the likes of Spotify and Deezer and other music apps. I think the way that they are utilizing the data to create a whole new experience for us as customers, I think it's incredibly relevant and exciting, but yeah, are there any customers that you want to? No, I think that's exactly it. You got the application development out there. I think you hit a really good point. The definition of an application has changed, right? I won't show my age, but for us, applications initially started as mainframes and green screens, that was the application. Then we moved to the client server model, and now the applications really are carrying them in our pockets, and it's those applications that can live in multiple spaces that are driving a lot of different changes in that. Another one we probably don't think about, but our kids think about it all the time, is Netflix. We think of that as a delivery mechanism of entertainment, it's actually a big data application. There's an entire recommendation engine sitting behind that piece of the information so that the definition of what is an application, and we won't mention the big G word, but that's what they are. The entire thing is actually a big data application. How about the relationship and the time we have remaining? You guys have been partnering for a while. I mean, you're kind of both arms dealers, so you do different things, so you're sort of partnering up, but maybe Mark, start with you from your perspective, what's the relationship like? So the relationship's great, and it's multifaceted right from the top down in the organizations. I think that it allows us both to have new conversations with our customers and our prospects. I think we are helping EMC to have an analytics conversation that maybe wasn't happening traditionally before. I think EMC are helping us massively by giving us an opportunity to talk about different infrastructure and different plumbing as it keeps being referred to here. I think it's great. I think from a joint development and engineering perspective, all the way down to the field, it's been extremely collaborative, it's been really enjoyable. So it's deeper than just to go to market? Oh yeah. So what kind of engineering and development have you seen us share in that? So a little bit of it, right? So obviously we have products in our system that are storage mechanisms as well as compute mechanisms. So we actually join engineer with Fortinworks to we test their new releases of their versions on our systems, understanding how new tools that are coming out, such as HDF, how those play into the model. So we test those to understand how they fit into our vision and can we actually utilize them to drive pieces. The other part for us is from a partnership perspective that's joined for us is being able to take a lot of the technical and business expertise from Fortinworks and the vertical spaces, telco financial services, and allows us to go in and have a business conversation. Not to be wrong, right? A lot of people look at us and it's EMC we're a technology company and we utilize the partnerships with Fortinworks to be able to jumpstart past the technology and understand the companies how they can actually change their business and how we can help them with that journey. And that's a big partnership for us. That ties in really nicely actually because we've got an industry solutions team inside of the business. And you mentioned before, Kerry, about focusing on the outcomes rather than the plumbing and the technology underneath. And what we do in partnership with EMC with this team is drive those conversations, take people from within our business who have come out of these different segments within the industries, telco, oil and gas, manufacturing, automotive and we can have a very relevant business discussion with them and then leave the technical stuff for a later discussion once they've started to understand the value that we bring as a partnership. Well, now with the Dell acquisition you just, the number of customers that you've increasing breadth and depth. So that's going to be interesting to see. EMC world's coming up first week of May. We'll be there with theCUBE. Michael Dell's giving a keynote. He'll be on theCUBE right after his keynote. We were just talking on Twitter. So we're really excited about that. Gentlemen, thanks very much for coming to theCUBE. Congratulations on Mark on this ecosystem growing and the success that Hortonworks is having. And Kerry, thanks very much for coming on theCUBE as well. My worries for having us. Appreciate it, thank you. All right, you're welcome, cheers. Keep right there, buddy. We'll be back with our next guest right after this short break. We're from Dublin, live. This is theCUBE, Boop Summit 2016.