 Live from Las Vegas, it's theCUBE. Covering Discover 2016 Las Vegas. Brought to you by Hewlett Packard Enterprise. Now, here are your hosts, John Furrier and Dave Vellante. Okay, welcome back everyone. We are here live in Las Vegas for HPE Discover 2016. This is SiliconANGLE Media's theCUBE, our flagship program. We go out to the events and extract the signals from noise. I'm John Furrier, my co-host Dave Vellante, our next guest. Dave Vellante, big data platform marketing at HPE Software, HPE for Enterprise. How's it going, great to see you. It's going great, how are you? Doing good. Big data is all the rage. Obviously this is on stage. We've heard a lot about machine learning. Yeah, machine learning is the new big data. It's AI. Write that down now and repeat it three times. Well, add AI to the list too, because that's coming right around the corner. But this basically highlights not why I should use big data. It's how, right? So everyone's in that conversation. So share your thoughts on what you're seeing for people's approaches, with how they're using the technology these days. What are some of the successes? Just share some color around. What people are doing with it? How they're instrumenting their business? I know it's different by industry, but share some data. Well, there's a couple real big drivers. Now that the big data has really landed and people are trying to figure out to your point, not just how do I store it and manage it, but what am I going to do with it? And at what velocity and to what end? And that's why at this discoverer, we really made a big bet across the company in shining neon lights around what we're referring to as applied machine learning. And it's not an if. It's really a when and a how because there's no other way to tackle the different forms of data that are hitting organizations. And machine learning, here's the interesting thing, is that AI machine learning, in one sense, it's not new. We have some proven technology that we've been proving and using and powering very serious applications like eDiscovery for over seven years. So if somebody just kind of woke up out of a time capsule, they'd go like, what changed? Because we were talking about this before and what has changed is pretty profound in that the technology itself, the computational assets that you can put at it, and the ability to be able to make it accessible. So it's not just for the few, but for the many, where we're really taking machine learning and AI and making it for the developer. That's a great point, a great point. It's not that the data scientist isn't still key, but it's now about going from fine China to everyday wear and having every developer and every application thinking about how you're going to use machine learning. Well, let's just capture that point. So one, access to more compute and or power is now available, that's changed, that's your point. But also the surface area for what people can see, the demand for new machine learning. You got iPhone apps, you got mobile apps, you have much more applications that could be machine learned or have applied machine learning to them, is a driver. I mean, there's more stuff to apply to and you got the horsepower. And you're bringing up what we've seen again and again, isn't it funny that the sophisticated stuff is done on the enterprise, but to actually show people you can think differently it happens on the consumer. So Siri, which people, they don't care it's machine learning or what's underneath it. Siri and the equivalents have just taught, and a grandmother, you know, that you can actually kind of talk to a computer and a computer can understand you and do some smart things for you. And that simple concept of that computers can learn translates all the way back to the enterprise. But as with a lot of things, it took a few consumer apps to kind of wake up that we don't have to lock in and be monolithic with enterprise computing. Well, it's always been hard on the enterprise side with the database, legacy, data structures, different schemas on databases. Consumer apps would just have a database. Now, with the consumerization of IT, you're seeing more complexity. So, is that why the consumers are talking? Is that what you're saying? Well, I think it's a necessity that this data is coming in and fundamentally companies have to be able to learn and act from it. If they don't, their competition will. And it is about speed and scale which has always been the challenge, but now it's speed, scale, and intelligence. And these new forms of data plus the structured forms are where you need the computers to do some of the lifting for you. And it's not to so much replace the human as augment the intelligence that's there. Now, the breakthrough for us with Haven on Demand was that all these technologies I talked about were not for the faint of heart. You needed significant investment, significant expertise. The economics were not there and that's why it's not prevalent. And what we changed by creating a cloud-based accessible through an API that any developer, any developer can now do face detection, now can do speech detects, now can do pattern recognition. That allows you to take this into any app. And that was something before Haven on Demand simply didn't exist. And I think it's gonna reset the bar. So you talked about developers. Robert Young-John talked about the importance of developers. It was one of the three pillars of your strategy. You're a platform. So obviously you got to get developers writing to that platform. The Nirvana, for many organizations, is you talked about needing data scientists and that's, you know, all of insights for a few. The Nirvana is actually putting it in the hands of the business users. What are you seeing, Jeff, in terms of that evolution, that sort of citizen data analyst, if you will, where are we in terms of your customers being able to get there? Well, I think it's going through a few phases. And phase number one is simply access, okay? Because a lot of this data was effectively dark and locked up and you needed to have a really good formal reason and had to be kind of institutionalized. So what has broken through on the last year or two or three, even this term of citizen analyst, which it's just speaking to the fact of just let the people have the goddamn data. And we put these nice terms around it so it seems like it's very thought through. But you had a pent up, huge level of frustration where people said, I want to be data-driven but I need that data to be accessible. I need it to be trusted and I need to be able to access it and react on and do things with it in a timely manner. Reading yesterday's news or yesterday's or last month's report, sometimes historical trending is useful but in most markets we're in, it's kind of like yesterday's newspaper, not particularly important. So that was the first phase where shadow IT appeared in all its glory and we saw the tableaus and clicks emerge that put tools into the hands of people on the edge that could utilize this. And I think admittedly the data center had to catch up to that. Now what's starting to happen is we're able to put in the infrastructures and the analytics that we can supply that. So you don't have the poor business analyst that just wants to know how they've been performing with their customer base, trying to suck out data and shove it in an Excel spreadsheet, we can do better than that. So in the value chain of things, you've got the data, you've got ways to store and act on the data which might be some kind of Hadoop distribution or whatever it is, some spark thing. And then you guys are taking the insights from that with Vertica and then feeding, you mentioned Tableau, Click, Pentaho's maybe another example where I've seen you guys participate in various shows that ultimately operationalizes against the GD data in the hands of the users, right? Is that the right way to think about it in terms of that flow of the value flow? Well, you, it's interactive though. That's the one thing that I would add to that. What do you mean by that? Well, when you bring the data out and do data discovery versus search, it's the value of the question. So we think of a lot of analytics as in the context just out of habit as reporting and therefore the answer is there. But when you actually look at companies that are data driven, it's about getting exposed to that data and then that will prompt questions. And you need to be able to ask that question back to the data. You need to be able to test hypotheses. You need to be able to do what's called data wrapping. I don't know if you've heard that term or if your users have. And that is bringing asymmetric parts of data together that may sit in very different locations. Some out on the public internet, some with your partners. We've heard data wrangling. Is that different than data wrapping? Yeah, data wrangling is a little bit, I think it's a funky way to talk about data mining, but the whole idea, get my data and get some sense to it and impose structure on that data so I can start to control it and start to extract out insights. And data wrangling is absolutely essential. Data wrapping says, I guess if I stay with the terms, it's after I've done my wrangling, now I have some core information, but now I want to know what about relationships with weather or economic conditions or geopolitical events or whatever. So I need to bring in other data into the mix and see if I can wrap that data with other insights. Do I start to learn more? Or can I create services that are going to deliver value? And it's all about building out your data but doing it in a fairly ad hoc fashion. And that's hard to do. That's your outcome, right? To validate the insights you're getting and then often to take action that you need to be able to do. And a lot of systems are designed opposite of that. They're about siloed relatively closed systems that will produce some sort of insight and done. And then maybe once a year, I'll move it into some grand warehouse, structured organize it and get some grander things. And now what we need is the ability of standing up at a whim and blowing away at a whim, these kind of computing. And let me explain more about that. The computing or the data sets or both? Well, both. Now it's not that I want to destroy my data, right? Because no data is being destroyed now. The whole idea, defensible disposal. The last time somebody disposed of data was probably 1927. They're data hoarders, it's a new cable show. People hold on to their data, okay? Sorry, General Counsel, it's just the fact. Yeah, it's just, you know, that smoking gun. But what we're getting at here is historically these systems, all of them, have been designed with a basic premmage of compute and storage. And the data goes with it as these interlocked welded together assets. And you designed around it. In fact, we put fancy words like architect. You had a analytics architecture that was presupposed. And you thought hard about what would be your peak usage and your utilization rates. And you had to make choices. Do I over allocate? Can I handle that monthly end report? How distributed is it? How reliable is it? And it was like building the Taj Mahal. You know, you measure twice, cut once. You were measuring a hundred times, you know? And then you were probably cutting a hundred times as you were building that thing out. And not much has changed today, even as we move to the cloud. And yes, the cloud will set up that underlying infrastructure, but you still have to specify it. You have to be able to tell your public cloud vendors, what do I need? And think about those resources. And the storyline hasn't changed much. So to preview what I think's coming, and what we're spending a lot of time thinking about is just ripping that whole concept apart of separating compute and storage and the data to allow you in a very agile fashion. You may say at two o'clock today, I want to be able to run this kind of analytic. I need this kind of compute. I need to pull together these data assets and I want to run this analytic. And then through automation, the ability to stand that up for what may be an hour. Get what you need and stand it down. So the ability to share that data, do that without impacting any of your other analytics. That's the big if. Because can organizations stand up that you say? It's like a developer with a sandbox. You want to basically flash the data into some sort of new environment. It sounds like composable infrastructure meets composable data. Well, it has some assets of that because it's all about bring that together. But the word I'm going to throw in there is temporal. That is my word of the day. The idea is can I do this in a temporal fashion so that I get what I need when I need it with very, very little friction or cost up or very little friction and cost down. Either what I'm trying to do or the other things I'm trying to accomplish. And that is the mecca holy grail that we're trying to go after. And I think that's going to be the big breakthrough. The assets are there, the underlying pieces. We can move things to the cloud pretty quickly. We can move data around her of access there, but we're still working. We still got a big foot in the old box of building things like we're building castles, right? You didn't build up a castle for an afternoon. You kind of worked for about 500 years. Now the good news was that castle, if you're lucky, would stand for another 500 years. Now we need things that are as brawny as castles, but we need them to go up in about two hours and we need to rip them down in about an hour. And that is the new castle building that's coming. So let's set up the big data conference come in the end of August, beginning of September. It's one of the events that we've done. We did the inaugural one. We've done it every year since. It's a great event. It's intimate, really a customer event. It's not a big vendor, you know, fast. It's all about the customer. Set up what's coming at BDC this year. Well BDC is, by definition, our biggest proudest event for our particular part of the business and HP software. And it defies some of the labels, right? Because is it a user conference? Well, yeah, you have a lot of people that are using and touching the technology. Is it a conference where you're gonna see strategic insight and sharing? Yes. The one thing it's not is a marketing conference. And I can say that because I'm a marketing guy. So we actually do not allow marketing to happen there. Every breakout session, everything that is delivered is delivered by one of two human beings. An engineer or a customer that is an engineer that's done it. It's about 1,500 to 2,000 people in Boston, August 29th to September 1st. We used to have in the middle of the summer and people still showed up. So now we've actually shifted a little bit so people can still enjoy that beach time. So the scheduling is pushed out a little later. And what's unique about this is it's focused on all people that are really passionate about pushing the edge around data. So these are the people that are architecting it, using it, driving analytics in their organizations. I mean, you guys had some great, I mean, we've been at that event since the deception and it's exceptional audience. I mean, we've had amazing conversations there. Again, they're like the DevOps guys. We met some guys from Facebook there. We met a bunch of people from all these hot companies. Zynga, Philo, all the web, all the web scale guys. Yeah. Well, and the names, and we will not disappoint this year. There may be a few names. There's one kind of ride sharing company. Some people may be familiar with begins with you. Lift. The other one. I always want to use lift in a keynote, not Uber, you know. Look at both there. But you're going to see some companies that are not just doing interesting things with data. Their businesses are defined around data. They are taking it and thinking about how they use it to the next level. And this isn't just about being impressed with somebody that's on a corner case thing. This is about learning and taking back to your organization, what can I do? And that's why people want to hear from Facebook and AT&T. It's over 60 breakout sessions. All of them run by people that are really doing it. The last thing is the busing, the busing aspect of it. And people may go, well, that doesn't sound very high tech. But the reason we hold this thing in Boston, besides being Boston, being a great town, is that that is our center of excellence for a lot of our engineering. So we put on buses, waves of buses, that are going from Cambridge down to the Westin that bring in our engineers. And they are the ones that staff and are part of the engagement zones that we have there. So you can go directly up to the people that are building this stuff using it and get the real baloney about what's going on. And the dates and those events just real quick. August 29th, September 1, three days. Boston, it's at the Western Waterfront. If you go to HPE, Big Data Conference, the early bird promotion ends in one week. And we all know a goodbye, a good promotion, makes us all happy, so you don't want to miss that. So check out HPE, Big Data Conference. It's different, and you will not be disappointed about this event. You will come back wiser. You'll be better looking, two inches taller. Great networking. And you'll be able to make contacts that maybe you haven't had before. You will also get a real sense of how we're setting the next stage for the next couple of years of advanced analytics computing. Awesome, Jeff, thanks so much for sharing the insight on theCUBE here, great data with theCUBE. Word of the day is temporal, and it could be a great CUBE gem. So thanks so much for that sound bite there. Temporal data, the future's coming. New way of doing things with Big Data with HPE platform. Be right back with more Big Data here in theCUBE. I'm John Furrier, Dave Vellante. We'll be right back. You're watching theCUBE.