 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. Hello everyone, we're back live here at theCUBE, live in Silicon Valley for Big Data Silicon Valley or Big Data SV. That's the hashtag, Big Data SV tweet. Go to crowdchat.net slash Big Data SV. This is theCUBE, our flagship program. We go out to the events, extract the signal and the noise. I'm John Furrier, the founder of Silicon Eagle. Show my co-host, Jeff Kelly, analyst at wikibond.org. And Bruno Aziza is here, CMO of Alpine Data Labs, CUBE alumni, great guests we've had on many times over his career and he continues to surprise and wow the crowd with props and stickers and giveaways here at the Stratoconference. Bruno, welcome back. Thanks for having me. You know, we love chatting with you. One, you're just a great guy and you're a tech athlete. Passion about it, you know what's going on and you're making things happen, mover shaker. That's what theCUBE's all about. So let's just dive right in. One, before we get into some of the commentary, want to just share with the folks, Alpine Data Lab, what's going on with the company? We have some props here, want to get into that as well. You gave away a Vespa to start up showcase at the Stratoconference. What's happening? What's happening on the ground here with Alpine Data Labs and the company? Well, thanks for having me here. There's a lot of things happening. First of all, we are announcing now a product that we called Alpine Course, which is bringing collaboration to data science. So this is about helping, not just data scientists being productive, but how you help data scientists reach out to business analysts and extend, if you will, their reach, as well as including executives. So we have this product that is essentially, think about it as a SharePoint of data science. It's a collaborative platform that allows you to do complex math, but allows you to reach out across the entire organization. There's no other company I know of that has a solution like that and it's been adopted by many customers. We had three of them yesterday in our session, Havis and Sony and Aurelia across online businesses, financial services, healthcare. It's really hitting a chord, this idea that the way we're gonna grow our space is to extend across the organization, not just make the data scientists productive. So that's what we're talking about here and we've been very happy about the reception. We did have a few events, you're right. We have some great stickers. We invite people to be part of the A team. So if you want to, you can do that. And then we have this great periodic table of operators that I can tell you more about. We did give a free Vespa two nights ago and there's actually a second Vespa you can win and I can tell you how you can do that later in our conversation. Great, so we were last time, we were visiting you guys in San Francisco. You guys have a new big office up here in San Francisco, great event. What's going on with the company? Obviously on the focus of Big Data, we're here at Big Data SB and Stratoconference. What's the update on the company in terms of where it's at, the product, and just the overall activity within the company? So when you were visiting us, we had just announced our Series B that was back in November where you raised 16 million from the guys at Sierra Ventures, Mission Ventures, and a few other investors involved there. And that really is going to allow us, if you will, to now scale our efforts. The first phase of the company was really about building very solid products, advanced analytics that is code free, that is on the web that has this technology called Incluster Analytics that allow people to essentially do massive machine learning at scale without having to pay a lot of tax in terms of infrastructure. So we have that. We're now starting to build our sales and marketing engine. And we have had amazing adoption from companies, like I said, Visa, and Barclays, and Sony. So companies you've heard of, as well as small startups like exactly here, for instance, well, I guess it's a sizable startup, but basically helping companies that are running into this issue where they have lots of data and they want to be able, not just do basic analytics, but start predicting where the market's going to be and will allow them to do that with our sales board. So who are your typical end users? Are they these really advanced data scientists, or are you trying to kind of bring data science down to earth and allow some mere mortals to start doing some of these advanced analytics? So we see this concept that we call the predictive enterprise, which really starts with the hardcore machine learning, data scientists, and so forth. The bigger base of user is really going to be the business analyst. In fact, we have, now we think we've established a good ratio where there's about one to about a hundred. So there's one data scientist of about a hundred people in the same company that would be involved. So those could be business analysts, those could be also executives like chief data officers or VP of analytics that want to be able to now take all the analytics assets and put them into one place. One reason might be, well, I just simply want to extend the power of my data scientist, but the other one, frankly, is also a policy problem, because we've had this proliferation of BI tools and proliferation of data platforms. Now we're starting to be the worldwide west a little bit. And so now executives want to rally everybody around the same platform so they could repeat the successes. And we find that our product is giving them what they want. So let's talk about the collaboration component. You released Chorus. Tell us about that. I mean, we hear about data science as a very collaborative discipline, but it's a challenge sometimes to actually make that collaboration happen. So what's your approach? Chorus, I'll do that. Yeah, and there's really three levels of collaborations. The first one is, how do you make data scientists themselves better productive? So, and the idea is you want to be able to harness the power of your data and unleash the innovation of your employees. And so once you have got a very efficient data scientist team, you want to now do the second level of collaboration, which is data scientists and business analysts. So if you have to show a business analyst and marketing set of lines of code, or you're writing R, or you're writing whatever language you've decided to choose, it's going to be really hard for them to participate in your process. And so when you do that, when you're asking your marketing analysts to understand code, in a way you're creating an artificial barrier to harness their innovation. So that's the second level of collaboration. The third one is what I was mentioning, the executives. And the executives, we have this great company that is talking about using our product as something to call it as a model factory. This idea that once you've created the model, the processes, you have the data, you know what a team looks like. You can take that model and apply it to any business across the organization. So that third layer, people are here interested in developing a culture embedded analytics inside their company. That's really hard, but that's where we're driving to. Start with the data scientist, data scientist to business analysts, and business analysts to the rest of the organization, executives being involved. So infusing analytics throughout the organization really. Which is, we hear about the data driven organization, but oftentimes it gets, I think in some of these early days of big data, it gets into science projects or kind of in the back room somewhere, it doesn't get out to the rest of the organization. It sounds like that's one of the problems you're trying to tackle. There is a little bit of commodity aspect, I think, if you think about, because I do hear a lot about data democratization and the data driven organization. And often when people say that, they think about more reports, better looking dashboards. And that's necessary, but the companies that we're working with, what they're interested in is actually, I want to play where the game's going to be at. In fact, I want to be able to master my own future. And that's where predictive analytics and the stuff that we work on are coming to play. So let's talk about the marketplace. Obviously Hadoop is at the mainstream Bruno. So I got to get your take on your view, because you've seen the landscape evolve over the past few years. Do you agree Hadoop is here to stay? I mean, it looks good. Hadoop is here to stay, but the focus of getting data and getting value out of data fast seems to be the concern here. What's your take on that? Yeah, absolutely. I thank you. There's no question. Hadoop is here to stay. And there's really three types of customers we see. We see a small contingent of customers that might not buy into your vision, what you said where basically they don't agree, they'll never touch it and maybe they'll die because they didn't take advantage of the opportunity. The polar opposite of that is people that are betting the farm on Hadoop. And they're doing everything in Hadoop as much as they can and they're running into security, real time, issues that they have to deal with with adding layers on the stack. The larger population is really people that are dabbling with both. So they have some bets on Hadoop and they might have some bets on MPP databases. And I think the challenge for the industry is how do we enable the people in the middle to have the access to Hadoop, the access to an MPP database? What we see in our solution is not so much the fact that we increase productivity but we can increase productivity across all stacks. So if you're running a regression algorithm with Alpine you can run it against Hadoop, turn around and run it against a green plan database and the business user doesn't really care. In one case on the fly we'll talk pig, on the other end we'll talk sequel. And I think that's where the industry is going is the business is back here. So that we've made the investment in infrastructure we need to now build solutions that are relevant to the business where they almost don't have to worry about what's under the hood I would say. So let's talk about the analytics market and the mind of the customer. When you guys talk to customers what are the big concerns that they have? What are they looking for? What are some of the business conversations and what are the business outcomes? What business outcomes are they looking for? Yeah, so the biggest scenario is that we see at least in our industry, we're spending a lot of time with business executives and it's churn, right? If you think about, and I think you and I were talking about earlier, what are they worried about? Well, I want to get new customers. I want to make sure that the customers that I have I can sell them more and it's typically easier to sell more to existing customers, I think there's research on that. And the third category is I don't want to lose customers and I want to be able to anticipate when I'm going to lose those customers. And so that's churn analysis. And customers we talked to, number one problem, churn analysis. Tell me how I'm going to prevent and what are the things that I need to do in order to prevent customer loss. The second one is things like product recommendation and they're startups that are focused on product recommendations only and that I think are pretty successful. We have of course a whole set of algorithms that people use in order to say, hey, if you came into my branch and you bought this and I knew that from you then the most likely next product you're going to buy from me is going to be this other one. There's a ton of other scenarios that are quite incredible on healthcare, for instance. You know, the ability to give a local patient specific treatment for their genomic footprint. You know, I mean, there's a lot of innovation being done there, but because you're able to bring all this data together and use a layer of application like ours that give you more intelligence and predict the outcomes before you even start the campaign, I would say. Jeff, what's your take on it? I mean, you've been following the big data space of $50 billion trend. What's your take on the analytics space? Well, I think there's a couple things. One is we need to make it easier for business analysts to do their jobs so that they don't have to be this hard to find data scientist. You know, there's even, what is the definition of data science? Even people don't even agree on that. But basically we've got to lower the level, lower the barriers to entry for somebody to start doing some advanced analytics. But I think the other thing around the analytics market is you have to start focusing more on solutions and less on the core tools. So Bruno, I'd love to get your take on this. Do you focus on some, whether they're vertical solutions or specific use case solutions, how do you kind of look at that? Or do you think it's enough to give them, give the data science really, here's the tool set you go ahead and use it or rather than focus on a particular business problem whether it's a vertical or a horizontal problem. So the short answer is today we are a horizontal platform and we cut across lots of different verticals. So we have customers in healthcare, financial services, online, really everywhere you can think of. But the language we use and the things that people resonate to is really what I would call horizontal solutions. So just like I was talking about, you know, churn analysis, product recommendation. Those are the types of terms that really are applicable across all types of industries. That's the direction that we're taking. I think if you want to be a vertical player, you have to have more than just a solution. You have to have the credibility. You have to have the support. You have to have a whole set of things that a software vendor like us, you know, would have to make a strategic choice in saying we're just going to go after healthcare. So we have partners for this. So when we go after a healthcare industry, we have partners that specialize in this financial services, same thing. But we try to do the best job we can in providing the tool set and the business layer on top of it that allows for a partner to go in and say, okay, Kaiser wants a, I don't know, the recommendation engine. We've done a lot of the recommendation engine work and you can customize on top of it for Kaiser specifically. And if you want to do that for Citigroup or Barclays or Visa, you can use the same recommendation engine, if you will, but you'll apply it for that industry. So tell me a little bit about how you go about kind of integrating with the larger ecosystem. We've been talking the last couple of days about how, you know, we think probably this year we're gonna see less necessarily product launches and product news and more partnerships and integration. Because really that's how you're gonna get to the enterprise and put together these solutions that can really start solving problems for businesses because it's not, you know, there's a lot of moving parts in the big end of the world from, you know, the hardware up through the visualizations and services. So how do you go about, I mean you mentioned Hadoop, you mentioned Green Plum, you've got to play with all these different technologies, you've got legacy databases, I'm sure you've got to take into account. What's your, I guess your partnership strategy and how do you approach that? Yeah, so there's two strategies there. The first one is with the ISVs, so, you know, Pivotal would be one of our great partners, for instance, we also partner with MAPAR and people in that ecosystem. And the idea there is that we want to, of course, help you with the best distribution that you want, but you know, some customers don't have distribution yet and we have a customer of ours that has a set of people working directly with Hadoop. And in a way, frankly, the problem we're solving for them is about Hadoop productivity. So at that layer, you know, from the stack, we try to be as agnostic as we can. Either you made a bet on the distribution or you haven't, and what we saw was a new productivity. Another ISV that's a great example is the partnership we have with Tableau. Because lots of customers using Tableau today, they want to do more than just reporting. They want to create data from the data they already have. So if you are doing some cluster analysis, for instance, and you're looking at it in Tableau, you might want to drop that data into Alpine, do some additional work using our algorithms and spit that back so Tableau can look at it. So on the ISV side, that's really our strategy. And then on the SI side, you know, services and so forth, we try to go after, you know, industry specific because we think they are probably the best and the closest people to the end customer and they're the most credible and probably the most valuable players at certain companies for industry specific. So we're not trying to build the industry muscle in our company. We really try to be the best enterprise software company we can for advanced analytics and collaboration. I really think this collaboration aspect is going to be big for lots of companies. And so another thing we've been talking about is this kind of transition from the talk in the market is less about the technology, more about the business value, right? And, you know, as that conversation happens, you're basically moving up the technology stack when you start talking about, you know, you're talking less about the dupe, you're talking more about the analytics. Have you seen that as well? And, you know, is that, how do you, what do you think we are in that conversation? I think this year is the year of productivity for big data. You know, I think we spent the last 24, 18 months or so focused on the data stack and what distribution and so forth. And I think for most companies, I don't know what you see but I think many of them have already made their bets. They might be switching cross players but, you know, they've decided either they go in by themselves or they've made a bet on distribution. So I think at the data layer solved. I think, sorry, I think the error value now is going to be at the top on top of that. And for that, you know, the business is back. The business is taking control over this because they've now realized there's a lot of power in being able to do analysis without samples, right? We joke in the company saying, if you're doing sampling, you're insulting your data because if you have paid for storing all that data and you have access to all that data, why don't you just analyze it all? But the reality is today there's a lot of friction in that, right? If you go to the traditional vendors that are predictive analytics vendors in MySpace, they'll ask you to move the data. They'll ask you to buy hardware to put it into a specific hardware environment. They'll ask you to download desktop applications which right away isolates your business analysts or data scientists. And we're breaking all those barriers, right? First of all, our solution's on the web. You just log in and start playing. Second, we have in-cluster analytics which you can basically play with the data where it lays. You can send instructions to the data without ever having to move it. So from an IT standpoint, we remove that friction as well. And I think adding the collaborative layer is now going to enable the business to engage the rest of the company in this process and it's very important that we do that. Otherwise, it'll just become a very specialized, nice industry, but I don't think it will explore the potential that it really has. Bruno, we've got to get on time here but Alpine Data Lab's doing great. No, it's great, we love having you, but we want you to explain the chart here. It's called the Data Science Periodic Table of Operators. That's right. Which is clever. I like the, and this says, if we're learning more, go to info.alpinenow.com slash predictive-enterprise. Yes, so. Explain the chart. And there's a QR code as well here so you don't have to type up all this. So basically the idea here is that. Follow it to the camera right there. The Data Science, where am I looking? Here, right here. So, and you can pick this up at our booth at 201 here. The world of Data Science might seem complicated to many of you, but what we thought we would do is we'd create this periodic table that looks very familiar for your in school and you studied chemistry and math. You probably referred to similar things like that before. And so what you'll find here is all the operators related to loading, exploring, transforming data that you can use as a reference. Now, if you go to the URL that's down here, you'll get an interactive version of that. So then you'll be able to essentially have it on one screen and if you want to know, well, what is, let's just pick a confusion matrix, you'll get the definition and so forth. Now the URL on the back will allow you to actually win a VESPA. I know that you wanted to get to that at some point. So, we are also- Cube host not allowed. So, we're building a new training program. And what we're doing is we're giving you access to this training program for free. And the first hundred people are going through it. After the hundredth person, we'll just pick the winner. So we're just trying to get a hundred data points to see how people are interacting with our training program. See if this is efficient. It's a little controversial because it's not a manual. See, this is not a hundred page manual like our competition will have you buy. This is a one pager and probably a training session and I'll take you three, four hours. So that's how we're trying to disrupt the market a little bit. Bruno, thanks for coming on the Cube. Alpine Data Labs, alpinenow.com is the URL. Analytics is hot. And again, getting value out of the data, the business conversations are pretty specific. The questions are known. How do you keep the customers happy? How do you delight customers? How do you get new customers? And how do you get information to keep serving those customers? And everything else is all just subordinate to that. Analytics are the key to the future and that's the business conversation and the business outcomes. This is the Cube. We'll be right back with our next guest. Live in Silicon Valley, this is the Cube.