 Live from the Fairmont Hotel in San Jose, California, it's The Cube at Big Data SV 2015. Okay, welcome back everyone. You're watching The Cube here live in Silicon Valley. This is our flagship program. We go out to the events, extract the civil noise. I'm John Furrier with Silicon Angle and The Cube. I'm excited to have two awesome guests here, Hortonworks and Pivotal because of the big news of the week, which we've been talking about basically since yesterday is the open data platform, Sean Connelly from Hortonworks and Leo Spiegel, Senior Vice President of Corporate Dial from Pivotal. Guys, stars of the show so far, fireworks exploding. Huge deal, I mean this is a big, probably one of the biggest announcements I've seen is pretty significant for the industry and we've been talking with you guys since the formation of the big data industry, watch Green Plum and EMC and the Pivotal, all that transformation, all that development from creation, okay, we're now years into big data. So welcome to The Cube and welcome to the segment. So I guess we'll start with Pivotal, obviously great success. Cloud Foundry has produced some great numbers. I was actually shocked, I thought it was mostly Green Plum but if you look at the totality of just what Cloud Foundry has done, huge success with Pivotal and now you guys are essentially open sourcing a lot of the investments you guys have made and Horton were essentially with Hadoop, the investments you guys made significant. So first, tell us about the announcement. You guys had a big event yesterday. What is the open data platform? We were just on IBM, they're making a huge bet. They're behind it. The entire industry's behind this event around standardizing Hadoop, what does it mean? Yeah, I mean, first and foremost, I mean, we think that the Apache world is fantastic. They do great work, all of the engineering, all of the development is key and we're not looking to change any of that. What we're really about is bringing together a bunch of companies to help accelerate the big data space and really enable people to have a common core that we standardize on so that all Apache bits coming from Apache and Sean will talk about that in a second, but that we all make sure that when we release our products, they're all using that standard core. In addition to that, what we really want is to enable all the companies to continue to innovate around that core. So lots of opportunity to create value and really meet the needs of the market. And a big thing you're gonna see with the open data platform is customers really voicing what they think is important and what they need to be successful. Because as we, as Pivotal, our journey is really about building this next generation of solutions for enterprises, helping them transform themselves and data and agile and our paths, our cloud foundry technology is really the hallmark of how we put these things together to help customers move in that direction. And data and data analytics in Hadoop is a really, really critical part of that. So I want you to talk back up for a second because there's been some critics out there saying, okay, just another industry partnership consortium, but the things are different. We're not living in a UNIX days and now there's post out there looking at the comparison to active desperation. But that's kind of the critics point. How do you answer that saying, hey, that's the naysayers of the critics, but talk about why that's different in this time where speed and agile and moving to the market is faster. How do you guys answer to that? Yeah, so a couple of observations. First of all, I think people should judge open data platform on our actions. And we just announced it, give us some time. I think we have a good track record. I mean, we took Cloud Foundry, which we built at Pivotal, spent a lot of money building. We put it in a foundation. Sam just joined as our CEO. I think we've come a long way. We have over 50 companies now building solutions on and around Cloud Foundry. And we intend to do the same thing with open data platform. And we're blessed to have great partners like Hortonworks and IBM and SAS and Splunk. And you know the list of 15 great companies. The other thing I would say is, from Pivotal's perspective, we've been all in an open source for a long time. Our heritage is spring source, which is kind of interesting because of the history that we have with Sean, obviously Cloud Foundry. And yesterday we announced that we're open sourcing our entire data platform. Over a billion dollars of investments made by basically EMC and VMWare, which we're giving to the community. And our plan is to, with Hawk and with Jim, apply for incubation at Apache. And we hope we're accepted. And then with regard to the Greenflum database, we'll do something different because it's based on Postgres. So we can't take it through Apache. But the point is these technologies will become available. Other people will be able to take that technology. And it's really about growing the industry, right? And enabling the industry to grow faster. If you actually look at the event yesterday, I think it's actually customer driven is kind of how we've been doing it. Who are the, just some of the names, and throw some names out there. Who's behind it? Obviously IBM's behind it. What are the partners behind it? Well, I think it's, if you look at what's demoed yesterday as an example, is Hawk on Hortonworks Data Platform managed by Mbari, right? That our journey together started last summer, effectively when we started collaborating around Mbari and those types of things. But really the goal here is, if you have a common core platform that not only runs the additional elements that we as Hortonworks focus on, but Hawk, Gemfire and other things that'll plug into that same chassis or IBM with big SQL or HP with Vertica or whoever, you provide the common substrate, then you enact a really solid choice, right? For end customers. Whether they're the GEs of the world or the others who participated in this announcement, a lot of this has been pushed by the end customers who at the end of the day, like a lot of what we may provide, but also like what a pivotal or an IBM provides and they want to get the best of both. So customer, how do you execute on that, right? Yeah, so let's just drill down on that. So people care about who's involved and how does it increase, accelerate the time to value, right? So let's talk about who's involved. So what are some of the names out there? Verizon, a major international telco, General Electric, SAS, Century Link. I have to go through my whole list of all these. Yeah, some players. Yeah, a lot of them are hitters and they're like little companies. Oh, a small company called Teradata. So I mean, these are some really important companies I think who are coming together. And it's really about, you know, doing good and accelerating the market. I mean, we think that at the end of the day, we can add value to the ecosystem and that's really what we're trying to do. So what are the customers looking for? Is it the standardization of Hadoop is a timing of releases? What, Sean? Yeah. And by the way, this is all being done in the open. Total, totally open. So that's I think a level of clarity I think it's worth putting a fine point on is the reality is it's sort of less about the production side, which Apache Software Foundation projects do that in spades, right? It's really about if you have the constant innovation that's happening in the projects and particularly across projects as well as commercial things that will plug in to the common platform is can you agree to ship a common version? So that way, Hawk and other technologies, SaaS and others can plug in and have a validation effort around a common version. So it's more of a consumption thing than a production thing, right? So open source works great for the production and the beauty is as consumers we're able to pick any and all types of versions and what we found was Pivotal's shipping a different version, IBM's shipping a different version, we're shipping a different version. So it's not the name of the component, it's actually version compatibility that helps downstream consumption, right? And we got briefed on Friday, so we'll do a little bit of the news and I asked the question, was Cloud Air involved in the deal? The answer is they're not involved as we saw some of the comments on the blogs yesterday. Were they invited? I mean, that's what people want to know. Absolutely. They were 100% invited, MapR was invited, all of the major companies that you would think about that are in this space were invited and they're welcome, full stop. I mean, we hope they'll join, we think it would be great for the industry and this is about openness, it's about co-opetition, right? I mean, IBM's a great example. I mean, we compete against IBM all the time with customers but we still find ways to work together on behalf of the industry and on behalf of our mutual customers. The other point I'll make is, as big data becomes more prolific, it's not unusual for big enterprises to have multiple versions of these technologies, right? And it becomes super complicated, right? They want to have a level of consistency. If you're an application vendor, you want to know that when you release the next version of your product, you don't have to test it on all these different distributions. Or migrate. Or migrate. So I mean, there's a lot of value that gets created in the ecosystem and to your point, things need to go faster. So why don't we find a way to work together to make things go faster? So I got to ask you guys, because you guys are both industry vets have been around the block and we're living in a new modern era and people who know me know I'm very, have been very skeptical of these big consortiums. When an open stack came out, I'm like, oh, this is just a big land grab, you know, Barney kind of deals. But then it really moved the needle in terms of people shipping code out there. And Cloud Foundry, same thing, seems like a land grab, but look at what's happened with the foundation. Now here, again, along the same lines, open data platform. Leo, talk about what you guys have learned because Cloud Foundry has been successful. You guys have got revenue behind, it's been some consolidation. What's different? Why are these new consortiums successful in this era now versus, say, back in the old days? Well, you know, it's interesting. I mean, if you go, I spent a lot of my time with CIOs. And if you talk to CIOs, what you hear is they want to know that the technologies they're betting on, they're not going to get locked into, right? At the same time, they want innovation on the top. And so with Cloud Foundry, one of the things we learned, and we really modeled ODP around it, was a common core. So lots of companies are taking Cloud Foundry, including IBM and HP, and they're building competing products to Pivotal, but we're all agreed that to call the product Cloud Foundry, it has to be based on that common core, full stop. I mean, version releases have to be synchronized, and it's super important. Versus owning a whole big monster of land grab core, as were you saying, that's the new kind of innovation. Yeah, I think that's the new innovation. And I think we did the same sort of thing here, and that leaves opportunities for companies to create value around the core. The other thing I think we learned with Cloud Foundry is, you know, there's been a lot of people saying, oh, it's all about the big corporations with all the money. But the reality is, we do have resources, but we want to leverage those resources in the open source space. And I think these foundations and associations are a mechanism to do that. They don't take away from innovation, they don't take away from developers, they don't take away from any of that. In effect, they can really accelerate it because we do have a lot of money. So why don't we use it together for the common good of the industry? And, you know, let's work- So talk about that money piece. So the theme is follow the money for our big data event this week. So are you buying market share? Are you donating market share? I think we're- Or both. I think we're more donating than buying. I mean, here are some of the things we're going to do. We're going to develop as part of ODP a standard CI suite, so that people can download that suite and test their applications on top of it. That'll cost some money, right? We're also going to work on eco-developing the ecosystem, right? We also want to work on education, right? There are, you know, one of the big questions when you go talk to enterprises is where are all these technical people who can deal with Hadoop, right? I mean, we're at such an early stage in this industry. So if you think of market amplification, we have the opportunity to actually amplify it into the industry. The feedback we get from customers all the time and the Wikibon guys have been tracking this is that customers want faster turnaround. It's not like the old days, the lag, the longer life cycles of software development, certainly shorter agile, what do you want to call this happening? Sean, I want to get your perspective. Rob Beard and you guys are both veterans of open source, we're all kind of first generation. We've seen the different eras of open source. What's different now in open source that makes these kinds of things viable and successful? I mean, what's different about this new generation? Yeah, today's reality is particularly in infrastructure platform software is it's done completely in open source, completely. That's the term. It isn't just this P with other pieces that if you want to pay for that value you can opt in. The reality is that slows down the adoption particularly on game changing infrastructure platform technology within the enterprise powering the cloud or what have you, right? And I think, so the expectations of consuming that type of model have changed dramatically over the past 10, 15 years. And what are the table stake tactics that need to be done in this market? Voting with code, is that kind of the standard? Well, I mean, you have to walk the walk, right? So at the end of the day the guys in our four walls, guys and gals in our four walls are, they code, right? That's what they do. We happen to contribute everything in our model to those other sources. Versus playing in there for PR reasons than having a proprietary little old spec. Yeah, exactly. That was kind of the old model. So, yeah, I mean, everyone wants a competitive advantage. Let's just be, people are in business to make money, it's not non-profit, right? Well, there's sort of accusations around open washing and things like that. I would say actions speak louder than words, right? So, give me an example is, and I was at the Pivotal event yesterday, the fact that Pivotals come to the decision to actually put hawk and gemfire and those types of things in open source, I'm telling you, I can guarantee you, that's not a marketing gimmick. That's not an open washing marketing gimmick. That's real. That's real. That's real stuff that's powering transactions and real solutions to that, right? So, sort of being called out in the intent, it's really up to these guys to kind of defend themselves, but I look at it and go on, no, you're actually bringing real code to the party. And it's cost too, so it's an asset for them that they're donating in this case. So, I mean, we were talking prior when we come on, the game is still the same in open source, you gotta win the audience and you gotta win the respect of the market. Yeah, I mean, there's no head face, a long game, right? No, and just look at infrastructure, software like .NET, who would've thought Microsoft would've done what they did, and oh, by the way, that runs on Linux as well, right? We're in a new era, guys, right? That's just how the technology's done. So, where does the value shift? Leo, you're the Corp Dev guy, you've done a lot of investments in startups, impressive track record. You look at the chess board, also have the inside baseball in this deal. What's the landscape look like? What is the value creation? Obviously, we've seen Pivotal evolve in VMware with Maritz, and we just interviewed Pat Gelsinger two weeks ago, and certainly network virtualization's a nice area that pass going after, and you got this kind of like middleware layer pass or Cloud Foundry, whatever they call it. Where's the value, where's the opportunity to create a competitive advantage for the big guys? You know, if you look around us, industries are completely being reshaped, right? Uber is killing taxi drivers. You know, Netflix killed Blockbuster. You can go from industry to industry. Right, Spotify, true cars, or man, there's all these guys that are just data-driven disrupting their perspective areas. Exactly, at the same time, you're gonna, you know, you have more people, I believe this is a true fact. It was told to me, so I haven't checked it, but I think more people have cell phones than toothbrushes on this planet, which is a little weird. So, I mean, think about the fact that you have more devices, at the same time, you have all these... What's the unit of things, floss? I mean, you know, wearables? Maybe, but then you have all these sensors, right? So then you have all these companies putting sensors on everything, from jet engines to people's socks to, you know, running shirts or whatever. All that information is gonna get put into basically a big data store, right? Call it a data lake. But the interesting thing is, what's the application or solution you build with that data, and how do you get value by building algorithms that can really transform the meaning of that data to change how your business works? You know, that's what companies want. That's what the future is. So analytics in memory, certainly Spark shows that. That's good testimony, right? The in-processor you see in Silicon, you're gonna see some new stuff coming around from the big guys there. But, you know, I was talking to Dave Vellante and Jeff Kelly this morning, and we were saying, you know, if this donation and its acceleration, which seems like to be the angle here, you guys are trying to move the needle faster for customers. If that's going to be the kind of, you know, table stakes, you guys are putting that in the middle of the pile there for everyone to use, that means the game is somewhere else. Are the stakes in the infrastructure software, where's the next big wave? So the question is, is it infrastructure software or what is that next big wave? And I guess let me clarify your other point is the ability to participate is broader now. 15 years ago, early days when I was at J-BOSS, it was talking to enterprises and the fact that GPL, Apache license, was not the boogeyman, right? Fast forward to today, the conversation I'm having with customers is we're betting our business on this technology. How do we go about changing our policy so we can participate, right? We have another initiative called the Data Governance Initiative, folks like Etna, Target, right? Merck, these are mainstream enterprises who want to participate in addressing data governance needs for their business and for their industries because they are hiring people from the workforce who have those skills. They just expect that the new wave of developers expect what they would do. So is this move in your era, you agree it's an acceleration, right? This is about accelerating value for customers. Where is the uptake going to come from? Developers, enterprise developers, pure, I'd say pure cloud, born-in-the-cloud developers. Where do you guys see the uptake on this? Ben, other, license? I think it's across all of those spectrums and I mean, the people who have the biggest advantage are the born-in-the-cloud companies, right? Because everything is instrumented. You can get whatever you need, you can have your, whether you do it on OpenStack or you do it on Amazon, obviously we think you should use Cloud Foundry since it's in the industry standard paths, but it gives you this abstraction layer to build your applications and run them wherever you wanna run them. But those folks have it easy. At the same time, you've got lots of big legacy enterprises that wanna do stuff that's third generation, but they have to live with the fact that they still may even be on the mainframe. And that's the interesting opportunity, right? Is to figure out how you can get all that information from all those systems into the application and create value. And I think that's, again, that's one of the big features. I think 10X the value above my area, 100X the value is in the apps and the solutions and the sooner those can get created, the easier you can make it, the more self-service you can make it. We gotta give props to Jeff Kelly two years ago, he called, the practitioners will get the value will be on that science equation. So you definitely made that call. Okay, so we got about a minute left, I wanna get one more talking point in it. The big splash in the pool is certainly ODP. When the water kind of levels out, you still gotta compete and win. So talk about the competition, the landscape out there. What's the competitive landscape look like for you guys in this new reality going forward? You know, first of all, I know you hate this answer, but there's so much opportunity, right? I mean, if you add up all of the revenue of all of the companies in this space, it's still a tiny fraction of the revenue that's being spent on data, right? I mean, so what we focus on, at least a pivotal, is how to make our customers successful, right? I mean, we did announce our data number yesterday, it's bigger than anyone's data number, it's all product. I mean, it's not to me about that. What it's really about is go out, create value for the customer and win. And this partnership that we've built, that's the other thing we didn't talk a lot about it. We are really partnering with Hortonworks. We think they're the right partner for us. They're gonna leverage our data technologies on the top, we're gonna leverage their Hadoop distribution. They're doing engineering for us. They're the best engineering company, I think, in the space. And I think that's kinda how I see the market moving, right? Let's figure out ways to work together to move the market forward and benefit the customer. If we do that, we can all have wonderful IPOs, we can all be successful. There's plenty of fruit on the tree. Plenty of fruit on the tree, that's not the problem. Don't fight that one tree. Make the pie big and I'll take a slice. The pie is pretty big, and Jeff Kelly sides it big. I asked Andy Jassy, or Dave Vellante asked Andy Jassy at Amazon directly, and he said it's over a trillion dollar, damn. I mean, it's trillions of dollars, and it's not like him. It's a rounding error, we're talking about, some of the numbers, you know. Enabling the, 100 million is a rounding number, right? I mean, come on. Enable them to create creative applications, move the needle in their business, drive more revenue, better margins. That's what I'm focused on, is enabling our customers to do it. So final question to wrap it up, Sean, what do you tell the open source community? What does this mean to them? Obviously, Apache's impacted. It's gonna be a new ripple effect throughout the industry. Obviously, there's a lot of value, developers care about distribution, they care about profits, and they wanna make money, but they wanna do the right thing. So what's the message to the open source community? So we've been consistent at what works in the fact that we do all our work in Apache projects according to the Apache way, right? So frankly, our role we feel in this is to ensure that that process is respected, period. That's not negotiable from our perspective. So what we wanna do is we wanna ensure that specific versions of various Apache projects as that core gets more defined around that can be consumed easily downstream by the solutions that are being built on that. So you get a little more predictability on the consumption side. As well as, and frankly, I brought up Data Governance Initiative and even the folks that the customers are participating in this, is to enable them to participate in the production process effectively. While Apache's been around a while, not everybody kinda knows that process and I think it's incumbent upon us to help folks through that journey. Leo, final question for you. Paul Moritz and the Brain Trust at Pivotal, do they see a visibility clear line of sight on the horizontally scalable enterprise? Yeah, I mean, I think Paul's been an amazing leader. I think we're very lucky to have EMC, VMware, and GE as our parents. And you'll see us make a series of announcements about what we're doing with customers. And I think how we're helping customers change their business and how we're enabling them to use our stack of technology that includes Cloud Foundry as our PAS, our data services, married with Cloud Foundry and our Agile services is really bearing fruit. And just watch our numbers, no forward-looking statements. We've got a lot to prove like everyone else. And I think it's- The word trillion was kicked around, so we don't expect that. There you go. In revised earnings segment. There you go. And we have our earnings call next week. Yeah, exactly. But I just, they're just- We're all getting in trouble with it. Yeah, if I just watch, watch what we do. I mean, I think at the end of the day, I think if we do what we think we can do, I think people will be impressed. Guys, thanks so much for coming. I really appreciated the big announcement. Congratulations to all the people involved. Big names, again, IBM and the Century Lincoln, among others, Ryzen, good market basket of leaders. So congratulations. Yeah, Jim and I, Infosys, I'm trying to think of a few more because we didn't get all 15 of them, but they're all important. Yeah, big player. I mean, yeah, big player. A lot of stakes, a lot of risk here in terms of big market opportunity move forward. So okay, guys, thanks for coming. This is theCUBE, extracting the signal from the noise you're hearing right here on the open data platform. We'll be right back after the short break.