 from San Jose, California, in the heart of Silicon Valley. It's theCUBE, covering Hadoop Summit 2016. Brought to you by Hortonworks. Here's your host, John Furrier. Okay, welcome back everyone. We are here live in Silicon Valley in San Jose for Hadoop Summit 2016. This is Silicon Angle Media's theCUBE, our flagship program where we go out to the events and extract the signal from the noise. I'm John Furrier, our next guest is John Mertek, the director of program management at the Linux Foundation. Welcome to theCUBE. Thank you, thank you for having me. We love the Linux Foundation. We love all the open source stuff because there's the communities now. I mean, there's so much open source going on. And as an older, old school dude, CS guy from the 80s, open source is so mainstream, such a tier one part of enterprises. It's great to see. So congratulations to all the work that you do. We support 100%. Just the update. What's the big news here from your perspective? Tell us. So from an ODPI perspective, we're announcing five distros that are ODPI runtime compliant. So back at Strata earlier this year, we announced our runtime spec, which focused on MapReduce, HEFS, and really core Hadoop, setting that base level standard within there. And we have five of our member distros, including IBM and Hortonworks, but also smaller distros, all to scale who's here at the event. Infosys, and also a smaller Russian distro, ArenaData, who have all went through the testing, went through the compliance, and they're now runtime compliant. So at Strata Hadoop, which coincides with Big Data SV, part of Big Data Week-R event with Strata, we had a panel on ODPI, and that was all the news. It was some controversy, some people aren't for it, some people are. Okay, that's open source. Right, right. No problem. But it was a big part of the announcement, right? So this is positive news. What's the bottom line impact for customers with ODPI, these certifications? Why is it important, and why should people pay attention to it? I know there's a lot of announcements going on at this event, so this is relevant, but why is this important? What's the milestones mean? It's, this is really the first time we're seeing market acceptance of what we're doing. We have companies that are putting their engineering dollars behind ensuring they're aligned with the standard. And if you look at it, there's basically five big distros out there. Two of them are ODPI runtime compliant. So your testing matrix, if you're an app vendor, has went to five to like four and a half. Who's not compliant? Cloudera, MapR, Amazon. Those are probably the key names, and there's a lot of other little distros, but this is actually the interesting other angle there. Taking away the focus from that, now a distro, now an app vendor that works successfully with IBM or Hortonworks, now it's an easy transition to Altascale. You know, it's not an uphill battle. It's a quick transition. It's a quick transition to Infosys. So you're opening your market up considerably with little engineering effort. That's a huge step in this ecosystem. And what's the benefit of the ecosystem? That means that there's reliance on the code. So talk about the dynamic of the bleeding edge and the stability of this because Linux went through this, I believe. So what's the, predict the future, if you will, given the history with Linux. So Linux had to have some stability and support, certainly Red Hat then, but now it does, what, 17 years and some God number, really big number. So 17 years, SLA is a huge number. That is a huge number, huge number. And frankly, innovation's going to continue here. That's not going to stop. The Apache Software Foundation has done a great job of championing that, but enterprises more and more, they're wanting to take this into real life production and they can't deal with that bleeding edge. They want to have a known target that when they encounter a distro, they know what to expect. There's things that are always there. It's really a support issue too for them, right? Yeah. Oh, integration key. Yeah, so now all of a sudden they're looking at ODPI compliance and saying, okay, I work with ODPI vendors. Now that's my list. And that's my focal point. And it's a good thing. So it's not just a rubber stamp kind of thing. The value is there and the risk would be that some would say is, well, they're not moving as fast as the rest of the market, but the vendors are motivated to lock step, just in cadence of that certification. So they do themselves a disservice by not being up to speed on hardened features. And exactly. And then the other half of it is, is it's helping all of these upstream projects. So we're, our certifications are basically getting the input from vendors of saying, what are you using out of this? What are the key interfaces? A lot of times these upstream projects really don't know. We're finding that out through our ecosystem through their use cases. We're codifying them into testing infrastructure. We're contributing that right back up to the projects. So now from the projects on down, is they need to release new versions? These APIs you're depending on? They stop breaking. John, give an example of a use case where you're saying that situation happened and that benefit was a direct result of ODPI. So, some of the early ones, a company I've been talking to, Splunk, who's one of our members, and they integrate really at that core, you know, MapReduce, HMS, you're on layer, right? And when they came to me, they said, you know, look, we're at Torhadoop and we've been told the story, this stuff has been stable for years. This shouldn't be a problem, but guess what? Every time that I'm getting a release, you know, this is the stuff that's not showing up in the release notes and I'm finding it through my QA cycles and now I'm having to kick this back to engineering to deal with these little inconsistencies and I'm fixing my product for things that should be stable but they keep breaking in the trailing edge. So that's kind of an interesting thing. It's not necessarily what's on the front that's breaking but it's the things that, hey, after we agree on it or stable, those start breaking because people stop paying attention to those. That's what really ends up being interesting. That's where our focus comes. That's basically you inherit a kind of undocumented non-QA feature so this ability solves that problem. Okay, so what's going on in the community? Tell us what's happening, the Linux Foundation and all the adjacent communities that you're involved in. What's that big update? It's a vibe. Well, you know, from the Linux Foundation, we're tons of collaborative projects we're working on in the networking space, IOT. I mean, that's just growing. It's growing and where we bring to the table is continues to be amazing. There's a ton of excitement what we're doing around here in Big Data. You know, I've been bouncer on the show for like crazy. We have a booth over there that's been going nuts. The sessions are packed. The sessions are packed but just really everyone that I go to and I just start talking the message of boy, is multi-distribution support a challenge for you and they say yes. And I tell telling our story, they resonate. They're like, where do we sign up? I mean, this is just no-brainer. It's a headache, it's basically aspirin, day one. Yeah, yeah, so it's been fantastic. I've loved talking with vendors. I've loved talking about their stories. And I think we're going to have quite a few members that are just going to come from this, if not just people participating in our community. And that's really what we're about. How can we get this voice to herald it together and just help make Hadoop easier for everyone? Awesome, well thanks for coming on and spending some time, John the Cube. Really appreciate it. Thank you. We'll see you up at your next event, to LinuxCon and all the big events coming up. You watching the Q here live, day two coverage here at Hadoop Summit 2016. We'll be right back with more coverage after this short break. Thank you.