>> Live from Barcelona, Spain, it's theCUBE. Covering KubeCon + CloudNativeCon Europe 2019. Brought to you by Red Hat, the CloudNative Computing Foundation and ecosystem partners. >> In mid-2014, announced the world, coming out of Google led by Joe Beda, sitting to my right, Brendan Burges and Craig McLucky, all Kube alumni. Kubernetes, which is the Greek for governor helmsman or captain and here we are, five years later at the show. I'm Stu Miniman and this is theCUBE's coverage of KubeCon + CloudNativeCon here in Barcelona, Spain. Joe you've got your title today is that you're a principal engineer at VMware of course, by way of acquisition through Heptio, but you are one of the people who helped start this journey that we are all on Kubernetes, thanks so much for joining us. >> Yeah, thank you so much for having me. >> Alright, so, the cake and the candles and the singing we'll hold for the parties later. We have Fippy and the gang have been watching our whole thing, for people who don't know there's a whole cartoon, books and stuffed animals and everything like that. Joe, when you started this merchandising, that was what you were starting, no. In all seriousness though, bring us back a little bit give us a little bit of historical context as to we've had you on the program a few times but yeah, here we are five years later was this what you were expecting? >> I mean when I remember Craig and Bren and I sitting around and we're like hey, we should do this as an open source project This is before we got approvals and got the whole thing started. And I think there was, like an idea in the back of our head, of like, this could be a big deal. You dream big a lot of times and you know that there's a reality and that it's not always going to end up being this. And so, I don't think anybody involved with Kubernetes in the early days really thought it was going to turn into what it has turned into. >> Yeah, so when we look at open source projects, I remember back a few years back, it was like to succeed you must have a phoenetical dictator that will make sure the community does this or wait we don't want too much vendor we're just going to let the user community take over and there's all these extremes out there, but these are complicated pieces. The keynote this morning the discussion was Kubernetes is a platform of platforms it's like I've got all of these APIs and by itself, Kubernetes doesn't do a lot. It is, what it enables and what things put together, so walk us through a little bit of that the mission, how it changed a bit and a little bit of the community and we'll go from there. >> Yeah, I think so early on one of the goals with Kubernetes from Google's point of view was to essentially take a lot of the ideas that had been incubated over about a decade, with respect to Borg and other things and so, a lot of the early folks who got involved in the project and worked on those systems and really bring that to the outside world as a way to actually start bridging the gap between what Googlers did and what the rest of the world did. We had a really good idea of what we were looking to get out of this system and that was widely shared based on experience across a bunch of relatively senior engineers. We brought in some of the Red Hat folks early on Clayton Coleman and some of the other folks who are still super involved in the project. I think there was enough of an understanding that we looked and said okay we got a lot of work to do let's just get this done. So, we didn't really need sort of the benevolent dictator because there was a shared understanding and we had senior engineers that were willing to make trade-offs to be able to go and move forward. So that I think was a key bit of the success early on. >> Alright, so you talked, it was pulling in some other vendor community there. Talk a little bit about how that ecosystem grew and when was user feedback part of that discussion? >> Yeah, I mean, when you say we pulled in the vendor we pulled in people who worked for vendors but we never really viewed it as, there was really from the beginning this idea of well what's good for the project? What's going to actually create sustainability and for the project, sort of project over vendor is really something that we wanted to establish. And that even came down to the name, right? Like, when we named the project, we could have called it Google XYZ or some sort of XYZ but we didn't want to do that because we wanted to establish it as an independent thing with a life of its own. And so, yeah, so we wanted to bring in those external ideas and I think early on, we did have some early users, we did listen to them but it really resonated with folks who could actually see where we were going. I think it took time for the rest of the world to really catch on with what the vision was. >> OK, when we look at today, there's a lot at the show that is on top of or next to or with Kubernetes it's not all about that piece. How do you balance what goes in it versus what goes with it? One of my favorite lines last year overall, was from you, saying Kubernetes is not a magic player it is not the be all and end all it is set with very specific guidelines. How do you avoid scope creep? As engineers it's always like, I don't know, we know how to do that piece of it better. >> So when we started out the project we didn't actually have a governance model. It was just a bunch of engineers that sort of worked well together. Over time and as the project grew, we knew that we needed to actually get some sort of structure in place. And so a bunch of us who had been there from the start got together, formed a steering committee, held elections. There's a secret architecture that we formed and these are the places where we can actually say what is Kubernetes what is Kubernetes not how do we actually maintain sort of good taste with how we actually approach this stuff and that's one of the ways that we try to contain scope creep. But also, I think everybody realizes that a thriving ecosystem whether officially part of the CNCF or adjacent to it, is good for everybody. Trying to hold on too tight is not going to be good for the project. >> So, Joe, tremendous progress in five years. Look forward for us a little bit. What does Kubernetes 2024 look like for us? >> Well a lot of folks like to say that in five years, Kubernetes is going to disappear. And sometimes they come at this from this sort of snarky angle. (chuckles) But other times, I think it's going to disappear in terms of like it's going to be so boring, so solid, so assumed that people don't talk about it anymore. I mean, we're here, at something that the CNCF is part of the Linux Foundation, which is great. But how often do people really focus on the Linux kernel these days? It is so boring, so solid, there's new stuff going on, but clearly, all the exciting stuff all the action, all the innovation is happening at higher layers. I think we're going to see something similar happen with Kubernetes over time. >> Yeah, that being said the reach of Kubernetes is further than ever. I was talking to this special interest group looking at edge computing and IoT people making the micro-cage version of this stuff when the team first got together, I mean, is you must look at and said there were many fathers, many parents of this solution, but, could you imagine the kind of the family and ecosystem that would have grown out of it? >> I think we knew that it could go there I mean, Google had some experience with this, I mean When Google bought YouTube, they had a problem where they had to essentially build out something that looked a little bit like a CDN. And so there were some examples of sort of like, how does technology, like Boar, adapt to an Edge type of situation. So, there was some experience to borrow we definitely knew that we wanted this thing to scale up and down. But I think that's a hallmark of these successful technologies is that they can be used in ways and in places that you really never thought about when you got started. So that's definitely true. >> Alright, Joe, want to give you the final word the contributors, the users, the ecosystem community, what do you say with five years of Kubernetes now in the books? >> I just want to send a huge thank you to everybody who made it happen. This is, it was started by Google it was started by a few of us early on. But, we really want to make it so that everybody feels like it's theirs. A lot of times Brendan Burns and me and Kelsey wrote a book together and I'll do signing and a lot of times I'll sign that and I'll say thank you for being a part of Kubernetes. Because I really feel like every user everybody who bets on it, everybody who shares their knowledge, they're really a big part of it. And so thank you to everybody who's a big part of Kubernetes. >> All right, well, Joe, thank you as always for sharing your knowledge with our community >> Thank you so much. >> We've been happy to be a small part in helping to spread the knowledge and everything going on here, so congratulations to the community on five years of Kubernetes and we'll be back with more coverage here from KubeCon + CloudNativeCon 2019. I'm Stu Miniman and thanks for watching theCUBE. (upbeat music)

>> Let me check. (uptempo orchestral music) (uptempo techno music) >> Live, from San Francisco, it's theCUBE! Covering Red Hat Summit 2018. Brought to you by Red Hat. >> Hey welcome back everyone, we are live here with theCUBE in San Francisco, Moscone West, for Red Hat Summit 2018. I'm John Furrier, co-host of theCUBE, with John Troyer co-host, analyst this week. the TechReckoning co-founder. Our next two guests are Ashesh Badani, vice president and general manager of OpenShift Platform and Alex Polvi, CEO of CoreOS, interview of the week because CoreOS now part of Red Hat. Congratulations, good to see you again. Thanks for coming on theCUBE. >> You're welcome. >> So obviously this is for us, we've been covering both of you guys pretty heavily and we've been commenting very positively around the acquisition of CoreOS. Two great companies that know open-source, pure open-source. You guys got the business model nailed down, these guys got great tech. You bring it together. So the first question is how's everyone doing? How's everyone feeling? And where's the overlap, if any and where's the fix? Explain the true fit of CoreOS. >> I'm going to start Alex, you want to jump in after. We're very excited right, so when we first had interactions with CoreOS, we knew this is going to be a great fit. The conversation we had earlier, both companies delivers in open-source, delivers in the mission center to take us forward regard to Kubernetes, as the container orchestration engine, and then being able to build out value for our customers around it. I think from our perspective, the work that both CoreOS did in advancing the community forward but also the work they've done around automation or their upgrades, management metering, charge back and so on. Being able to bring all those qualities into Red Hat is incredible. So I think the fits been good. It's been three months, I'll let Alex comment some more on that but we've been doing a lot of work from integration perspective around engineering, around product management. At Red Hat Summit this week, we reveal details around some of the converged road maps, which I can talk about some more as well. So we're feeling pretty good about it. >> Alex, your reaction. >> Yes, it's been three months. If you've studied CoreOS at all, you know everything that we do really centers around this concept of automated operations. And so by being part of Red Hat, we're starting to bring that to market in a much bigger and faster way of really accelerating it. The way the acquisition are really successful is either mutually beneficial to both companies and they accelerate the adoption of technology and that's definitely happening. We had the announcement yesterday with Red Hat CoreOS around the Linux distribution. Last week, we did the operator framework. It was very central to the work that we've been doing as part of CoreOS, and then as companies in a lot of ways is being part of Red Hat for three months now. This is what our company would have looked like if we ever just another 10 years along or whatever very similar, we're like a mini Red Hat, and now we're leaped ahead in a big way. >> And you guys done a good work. We've documented on theCUBE many times, and we were in Copenhagen last week. Now covering the operating framework but I want to get your reaction. You guys did a lot of great work on the tech side obviously, you can go into more detail but we've always been saying on theCUBE. If you try to force monetization in these emerging markets, you're optimizing behavior. And this was something that's gone on, we've seen containers. It's been well documented obviously what's happened. It's certainly a beautiful thing. Got Kubernetes now on top working together with that. If as an entrepreneur out there that are building companies. If you try to force the monetization too early, you really thinking differently. You guys stay true to it. Now we've got a good home with Red Hat. Talk about that dynamic because that was something that I know you guys faced at CoreOS and you've managed through it. Tempted probably many times to do something. Talk about the mission that you had, staying true to that and just that dynamic. It's challenging. >> Yeah, as we set out to build a company in general, there are really three operating principles. There is build a great technology to solve our mission which is to secure the internet through automated operations, build a great place to spend their days which is really about the people and the culture and so on. Why are we doing this, and the third was to make it sustainable and by that I mean to build their own money fountains, building out of the middle of our campus. And so by joining Red Hat it's we have a money fountain sitting there. (laughing) It's spewing off a ton of cash flow every single quarter that allows us to continue to do those first two things in perpetuity, and that third one is something every company needs in order to continue to execute towards the mission. And the thing that's so awesome about working with Red Hat is we're very much aligned and compatible. Red Hat's mission isn't exactly the same thing we are working but it's definitely compatible. It's like Apache and GPL are compatible. It's like that type of compatible. >> You both believe in open-source in a big way. Talk about the Red Hat perspectives. Now you got like a kid in a candy store. Openshift made a big bed with Kubernetes. You see now, you have the CoreOS, how has it changed in Red Hat internally? Things moving around actually accelerates the game a bit for you guys, and you're seeing new life being pumped into OpenStack. You're seeing clear line of sight with Kubernetes on the app side. We were just at KubeCon. A lot of people are pretty excited. There's clear lines of sight on what's defacto. What people are going to build around, and also differentiate. >> Right, so I'll start off by saying I really hope our CEO, Jim Whitehurst doesn't see this interview but if it goes off in terms of money factor. I'm currently make budget request. I think I know what's going on. >> Balance sheet, cashless now. It's in the public filings. If I see a fountain of money spewing off the thing, >> The ability to reinvest. >> This is a really good fit. (laughing) The way to say this, they have a great business model. >> Yeah, yeah. >> Some of us will make money, some of us will spend the money. Some of us will spend the money, it will work out well. (laughing) >> It's a great win. It's a great win. It obviously accelerates the plans. The commercialization is already there with Red Hat. This is just a good thing for everybody but the impact of you guys accelerating, just seeing OpenShift. You can boil it down to the impact of Red Hat. What is the impact? >> So in all seriousness, I think the focus for us really has been about there is so much complimentary work that's been going on with the CoreOS team that we're bringing into OpenShift, and to Red Hat in general that accelerates everything that you're seeing. You saw some amazing announcements happen this week with regard to our partnership with Microsoft and getting OpenShift out and Azure, and joint support offering. The work we're doing with IBM to get IBM middleware as well as IBM Cloud Private support integrated with OpenShift. The work that Alex referred to around automation, being able to bring that to our customers. We see all the excitement around that front as well so we want to take all Techtonic work that has been going on at CoreOS, then move that to OpenShift. Carry forward the community that CoreOS built around Container Linux, and actually inject a lot those ideas into that Linux, our flagship technology. Bring that passion and energy to bear as well, and then carry forward a lot of the other projects that they have. For example, the Quake Container Registry, that's extremely popular. Carry that forward, support our customers to use that both stand alone integrated with the OpenShift platform. Other projects like FCB that Alex has been talking about which is the underpinnings of Kubernetes plus running worldwide. So all of those things, we can bring forward, and then all the advancements that were made in place by CoreOS as they're working towards their money fountain, just plug that right into it. >> And just as a point of reference, Brendan Burns flew in yesterday. Microsoft Build is going up so he left their own conference to come down here. >> As did Scar Guthrie, right? >> That's a great testament. This is the testament. They're coming down, really laying down support. This is a real big deal. This is not a fake deal, it's real. >> And so I want to talk a little bit about specifics of the timeline, the road maps. Sometimes with these mergers or acquisitions, it's well the technology will be incorporated at some point, and then it goes away to die and you never see it again. And then the people all leave, and then you ask what was going on. But here, you actually have, I was great. You were talking to me. You have some specific timelines and we'll start to see some of the Techtonics Stack in OpenShift fairly soon. >> Yes, absolutely so the acquisition was announced three months ago and we said at that time that by Red Hat Summit, we'll lay out for you a road map and so we're now starting to do that. We put out release of some materials around some details with regard to how that's coming out. We have detailed sessions going on at Red Hat Summit around the integration plans between Red Hat, OpenShift and CoreOS with a few specific areas with regard to OpenShift. You'll start seeing the earliest versions if you will of the work that's being done. This summer, we'll deliver the full road map to you there by the end of this calendar year. With regard to, for example pieces like the Quake Container Registry that's being made available and being sold now as we speak. Customers can go get that, and we want to make sure no customer is left behind. Right, that's a principle we put out. And with regard to supporting any existing customers on Techtonic or the Container Linux space, we're doing that as we're working to integrate them into the Red Hat portfolio. Can you talk a little bit about the decision for Red Hat's atomic coast and Container Linux? Now re-named again, CoreOS. That was one of the seminal inventions that you all made as you started the company. I think it had some brilliant ideas again about security and the operational aspects but can you talk about some of those technologies and the decisions made there? >> Yeah, like I said, the acquisition of CoreOS Red Hat was about saying look what can we take that CoreOS has been doing to accelerate both work and community but also what could be doing to deliver this technology to customers. So the goal was we'll take all the atomic and the word that's been going on there have that be superseded by the work that's coming out of CoreOS Container Linux carry the community forward. Release a version of that called Red Hat CoreOS and in its initial form make that actually an underlying environment to run OpenShift in. Okay so for customers who want the automation that Alex talked about earlier. They made that available both at the underlying platform. Make it available in OpenShift platform itself via the work that's come from Techtonic, and then ultimately, Alex will talk about this some more through operators. So trusted operations from ISP or third party software that would run on the platform. All right so now if you will, we'll have full stack automation all the way through. OpenShift also support Red Hat Linux, a traditional environment for the thousands of customers that we have globally. Over a period of time, you should expect to see much of the work that's going on Red Hat CoreOS find its way into it as well. So I think this just benefits all around for us both in the near term as well as long. >> And Red Hat Container certification, where does that fit into all this? >> Yeah, a great question, so what we announced maybe was, actually was two years ago was a Container certification program. Last year, we spent some time talking about the health of those containers, and being able to provide that to customers. And this year, we're talking about trusted operations around those containers. That carries forward, we've got hundreds of ISPs that have built certified containers around it, and now with the operator framework, we've had, I think it's four ISPs demonstrating previews of their operators working with our platform as well as 60 more that are committed to building ISP operators that will be certified again. >> So people are certified in general, pretty much. I think we're very excited. The fact that we went to KubeCon last week, announced that the operating framework have been based on the ideas that the CoreOS team has been working on for at least two years. Making that available to the community and then saying for the ISPs that want a path to market. Going back to the money fountain again for the ISP that want to pass through market which is pretty much all of them. We also have the ability to do that so give them an opportunity to make sure that as wide as possible some adoption of the software at the same time help with commercialization. >> Can you guys share your definition of operator because I saw the announcement but we we're on a broader definition when we see the DevOps movement going the next level. It's all about automation and security, you mentioned that admin roles are being automated in a way to see more of an operator function within enterprise and emerging service providers. So the role operator now takes on two meanings. It's a software developer. It also is a network operator, it's also a service, so what is that, how do you guys view that role because if this continues, you're going to have automation. More administrator is going to be self healing, all this stuff is going to go on. Potentially operations is now the developers and IT all blurring together. How do you guys define the word operator in the future state? >> Well I know the scenario of great interest to you. >> So operator is the term for the piece of software that implements the automated operations. And so automated operations, what is that? Well that's what sets apart, the way I think about it is what sets apart a cloud provider verses a hosting provider. It's a set of software that really runs the thing for you and so if we're going to get into specific Kubernetes lingo, it would be an application specific controller. That's a piece of software that's implements the automated operations. And automated operation is a software that gives you that simplicity of cloud. It's at the core of a database as a service. It's both hosting but also automated operations. Those two things together make up a cloud service and that software piece is what we're decoupling from the hosting providers for the first time and allowing any open-source project or ISP brings the simplicity of cloud but in any environment. And that's what the operator is a piece of software that actually goes and implements that. >> So a microservices framework, this fits in pretty nicely. How do you see obviously? >> Microservices, there's all these terms. Microservice is more of an architecture than anything but it's saying look, there's these basic things that every operations team has to go and do. You have to go and install something, you have to upgrade it, you have to back it up, when it crashes in the middle of the night, get it going again. A lot of these things, the best practices for how you do them are all common. There's no ingenuity in it. And for those things, we can now because of Kubernetes write software that just automates it, and this was not possible five years ago. You couldn't write those software. There were things like configuration management systems and stuff like that that would allow companies to build their own custom versions of this. But to build a generic piece of software that knows how to run application like Prometheus or a database or so on. It wasn't possible to write that and that's what the first four or five years of CoreOS was is making it possible, that's why you saw all these mat and new open-source projects being built. But once it was possible it was like let's start leveraging that. You saw the first operator come out about a year ago, and I think it was our ATD operator was the first one, and we started talking about this as a concept. And now we're releasing operator framework which is from all the learnings of building the first couple. We now made a generic, so anybody can go and do it, and as part of Red Hat, we're now bringing it to the whole ISP ecosystem. So the whole plan to make automated operations ubiquitous is still well underway. >> I'd love to extend that conversation though to the operator, the person. >> Right. I think you and your team brought the perspective of the operational excellence right to the table. A lot of cloud has been driven by the role of developer and DevOps but I've always felt like well wait a minute operators the people who use to be known as IT insisted they had a lot to bring to the table too about security and about keeping things running, and about compliance and about all that good stuff. So can you talk a little bit as you see the community emerging, and as you see all these folks here. How do you talk to people who want to understand what their role is going to be with all this automation in keeping the clouds running? >> Computers use to be people too. (laughing) But we're not going to completely automate away everything because there's still parts of this wildly complex system that justifies whole conferences of thousands of people that require a whole lot of human ingenuity. What we're doing is saying let's not like do the part that is the fire drill in the middle of the might that keeps you from making forward progress. The typical role of an operations person today is just fighting fires of mundane things that don't actually add a lot of value to the business. In fact, this guy is difficult because you only get brought up when things are on fire. You never get an praise when things are going well. And so what we want to do is help the operations folks put out those fires like the security updates. Let's just roll those out automatically. The way you do those across all organizations does not need to be special and unique but they're really critical to do right. >> Well it's just automate that stuff away and let the operations team focus on moving the business forward. The parts that require the human spirit to actually go and do, and if we get to a point where a CEO of a company is like, wow, I can not come up with a new vision for this imitative 'cause my operations team are just so fast at influencing them. Then we have to start worrying about operations people's job but I don't see that happening for a very long time. >> And no one is going to be sitting around twiddling their thumbs either. >> Let me just extend that point a little bit. The whole point of operators is to encapsulate human knowledge that ISPs have and bring that in the platform and automate it. So the challenge that we've had is an operations person is required to know a lot about a lot. So the question then really is how can we at least take some of what's already known by people and be able to replicate that and that allows for every one to move forward. I think that's just forward-- >> Well, there's a bigger picture beyond that, so I agree but there is also scale. With cloud, you have scale issues. So with scale automation is a beautiful thing 'cause the fire has also grown exponentially too so you can't be operating like this. Scale matters, super. >> The reason that this stuff was invented at Google initially was not because of Google's high career per second. Is that they were, to build the application they're building required so many servers that you couldn't hire enough operations people without writing software to automate it. So they were forced to custom design the system because they had so many servers to run to build the software that they wanted to build. And other companies are just now getting to that point because every company is going through a digital transformation. They have to have thousands of servers just to run their applications. There's no way you're just going to hire the operations staff to go and do it all by hand. You have to write software to turn the operations people into mech warriors of running servers. You need to wrap them in automation in order to scale that. >> At KubeCon, she made a comment that all those operations folks at Google are software developers. >> Brand engineers. >> Brand engineering, so they're not Ops guys just pushing buttons and provisioning gear and what not. They're actually writing code. You bring up the Google piece, the other piece that we heard at KubeCon. We hear this consistently that this is now a new way to do software development. So when a former Googler went to work for another company, left Google. She went in and she said, "Oh my God, you guys don't do. "You don't use board?" To her, she's like how do you write software? So she was like young and went out in the real world and was like wait a minute, you don't do this? So this is a new model in software development at scale with these new capabilities. >> I think so and I think what's really important is the work we're doing with regards to an ecosystem perspective to help folks. So one of the top things I hear from customers all the time is this sounds fantastic. Everyone's talking about DevOps or microservices or wanting to run Kubernetes at scale. Do I have the skills? Can I keep up with the change that's in place and how do I continue going forward around that? So we announced at Red Hat Summit Managed offerings from let's say Atos and DXC where you've got goals to integrate us helping folks, or companies like Extension T systems. The CEO came and spoke today about the work we're doing with them to help connected cars, and those applications be rolled out quick and fast. I think it's going to take a village to get us to where we want to because the rate of change is so fast around all of these areas and it's not slowing down that we'll have to ensure there's more automation and then there's more enablement that's going on for our customers. >> So some clarity, can you guys comment on your reaction to obviously we've seen OpenStack has done over the years and now with well Containers, now Kubernetes. You seeing at least two ecosystems clearly identified. Application developers, cloud native and then I would call under the hood infrastructure, you got OpenStack. Almost it clarifies where people can actually focus on real problems that the Kubernetes needs. So how has the Container, maturation of Containers with Kubernetes clarified the role of the community? If this continues with automation, you can almost argue that the clarity happens everywhere. Can you comment on how you see that happening? Is it happening or is it just observation that's misguided? >> I think we're getting better with regard to fit for a purpose or fit for use case. All right, so if you start thinking about the earliest days of OpenStack. OpenStack is going to be AWS in a box, and then you realize well that's not a practical way of thinking about what a community can do a build at scale. And so when you start thinking about a Word appropriate use case for this. Now you start betting if you will, a set of scales, you set expectations around how to make that successful. I think we'll go through the same if we haven't already or even going through it with regard to Kubernetes. So not every company in the world can run Managed World call. DYI Kubernetes, don't many companies will start with that. And so the question is how do we get to the point where there's balance around it and then be able to take advantage of the work? For example, companies like Red Hat work for us was doing to help accelerate that path 'cause to the point Alex was trying to make is the value for them being able to keep up with the core release of Kubernetes? And every time a bug shows up to go off and be able to fix and patch it, and watch that or is the value building the next set of applications set on top of platforms. >> That's great, well congratulations guys. Thanks for coming on theCUBE. Appreciate the insight. Congratulations on the three months into Red Hat. Good fit, and enjoy the rest of the show. Thanks for coming on, I appreciate it. >> Thanks. >> Live from Red Hat Summit, it's theCUBE's coverage here of Red Hat and all the innovation going on out in the open. We're here in the middle of, we open the floor with Moscone West with live coverage. Stay with us for more after this short break. (uptempo techno music)

>> Announcer: Live, from Copenhagen, Denmark, it's theCUBE. Covering KubeCon and CloudNativeCon Europe 2018. Brought to you by the Cloud Native Computing Foundation and its Ecosystem Partners. >> Well everyone, welcome back to the live coverage of theCUBE here in Copenhagen, Denmark for KubeCon, Kubernetes Con 2018, part of the CNCF, Cloud Native Compute Foundation, part of the Linux Foundation. I'm John Furrier with Lauren Cooney, the founder of Spark Labs, breaking down day two, wrapping up our coverage of KubeCon and all the success that we've seen with Kubernetes, I thought it would be really appropriate to bring on the cofounder of KubeCon originally, Joseph Jacks, known as JJ in the industry, a good friend of theCUBE and part of the early formation of what is now Cloud Native. We were all riffing on that at the time. welcome back to theCUBE, great to see you. >> Thank you for having me John. >> So, for the story, for the folks out there, you know Cloud Native was really seen by the devops community, and infrastructure code was no secret to the insiders in the timeframes from 2010 through 2015, 16 timeframe, but really it was an open stack summit. A lot of people were kind of like, hey, you know, Google's got Kubernetes, they're going to open it up and this could be a real game changer, container, Docker was flying off the shelves. So we just kind of saw, right, and you were there and we were talking so there was a group of us. You were one of them. And you founded KubeCon, and bolted into the, at that time, the satellite Linux Foundation events, and then you pass it off as a good community citizen to the CNCF, so I wanted to just make sure that people knew that. What a great success. What's your impression? I mean, are you blown away? >> I am definitely blown away. I mean I think the size and scale of the European audience is remarkable. We had something like slightly less than half this in Austin last year. So to see more than that come here in Europe I think shows the global kind of growth curve as well as like, I think, Dan and someone else was asking sort of raise your hand if you've been to Kubecon Austin and very few actually, so there's a lot of new people showing up in Europe. I think it just shows the demand-- >> And Dan's been traveling around. I've seen him in China, some events I've been to. >> Joseph: All over. >> He's really working hard so props to him. We gave him some great props earlier. But he also told us Shanghai is coming online. >> Joseph: Yeah. >> So you got Shanghai, you to Barcelona next year for the European show, and of course Seattle. This is a community celebrating right now because there's a lot of high fives going on right now because there's a lot of cool, we've got some sort of core standard, defacto standard, now let's go to work. What are you working on now? You got a stealth startup? Share a little bit about it. I know you don't want to give the details out, but where is it kind of above the stack? Where you going to be playing? >> Sure, so we're not talking too much in terms of specifics and we're pretty stealthy, but I can tell you what I'm personally very excited about in terms of where Kubernetes is going and kind of where this ecosystem is starting to mature for practitioners, for enterprises. So one of the things that I think Kubernetes is starting to bring to bear is this idea of commoditizing distributed systems for everyday developers, for everyday enterprises. And I think that that is sort of the first time in sort of maybe, maybe the history of software development, software engineering and building applications, we're standardizing on a set of primitives, a set of building blocks for distributed system style programming. You know we had in previous eras things like Erlang and fault tolerant programming and frameworks, but those were sort of like pocketed into different programming communities and different types of stacks. I think Kubernetes is the one sort of horizontal technology that the industry's adopting and it's giving us these amazing properties, so I think some of the things that we're focusing on or excited about involve sort of the programming layer on top of Kubernetes in simplifying the experience of kind of bringing all stateful and enterprise workloads and different types of application paradigms natively into Kubernetes without requiring a developer to really understand and learn the Kubernetes primitives themselves. >> That's next level infrastructure as code. Yeah so as Kubernetes becomes more successful, as Kubernetes succeeds at a larger and larger scale, people simply shouldn't have to know or understand the internals. There's a lot of people, I think Kelsey and a few other people, started to talk about Kubernetes as the Linux kernel of distributed computing or distributed systems, and I think that's a really great way of looking at it. You know, do programmers make file system calls directly when they're building their applications? Do they script directly against the kernel for maybe some very high performance things. But generally speaking when you're writing a service or you're writing a microservice or some business logic, you're writing at a higher level of abstraction and a language that's doing some IO and maybe some reading and writing files, but you're using higher level abstractions. So I think by the same token, the focus today with Kubernetes is people are learning this API. I think over time people are going to be programming against that API at a higher level. And what are you doing here, the show? Obviously you're (mumbles) so you're doing some (mumbles) intelligence. Conversations you've been in, can you share your opinion of what's going on here? Your thoughts on the content program, the architecture, the decisions they've made. >> I think we've just, so lots of questions in there. What am I doing here? I just get so energized and I'm so, I just get reinvigorated kind of being here and talking to people and it's just super cool to see a lot of old faces, people who've been here for a while, and you know, one of the things that excites me, and this is just like proof that the event's gotten so huge. I walk around and I see a lot of familiar faces, but more than 80, 90% of people I've never seen before, and I'm like wow this has like gotten really super huge mainstream. Talking with some customers, getting a good sense of kind of what's going on. I think we've seen two really huge kind of trends come out of the event. One is this idea of multicloud sort of as a focus area, and you've talked with Bassam at Upbound and the sort of multicloud control plane, kind of need and demand out there in the community and the user base. I think what Bassam's doing is extremely exciting. The other, so multicloud is a really big paradigm that most companies are sort of prioritizing. Kubernetes is available now on all the cloud providers, but how do we actually adopt it in a way that is agnostic to any cloud provider service. That's one really big trend. The second big thing that I think we're starting to see, just kind of across a lot of talks is taking the Kubernetes API and extending it and wrapping it around stateful applications and stateful workloads, and being able to sort of program that API. And so we saw the announcement from Red Hat on the operator framework. We've seen projects like Kube Builder and other things that are really about sort of building native custom Kubernetes APIs for your applications. So extensibility, using the Kubernetes API as a building block, and then multicloud. I think those are really two huge trends happening here. >> What is your view on, I'm actually going to put you on test here. So Red Hat made a bet on Kubernetes years ago when it was not obvious to a lot of the other big wales. >> Joseph: From the very beginning really. >> Yeah from the very beginning. And that paid off huge for Red Hat as an example. So the question is, what bets should people be making if you had to lay down some thought leadership on this here, 'cause you obviously are in the middle of it and been part of the beginning. There's some bets to be made. What are the bets that the IBMs and the HPs and the Cisco's and the big players have to make and what are the bets the startups have to make? >> Well yeah, there's two angles to that. I mean, I think the investment startups are making, are different set of investments and motivated differently than the multinational, huge, you know, technology companies that have billions of dollars. I think in the startup category, startups just should really embrace Kubernetes for speeding the way they build reliable and scalable applications. I think really from the very beginning Kubernetes is becoming kind of compelling and reasonable even at a very small scale, like for two or three node environment. It's becoming very easy to run and install and manage. Of course it gives you a lot of really great properties in terms of actually running, building your systems, adopting microservices, and scaling out your application. And that's what's sort of like a direct end user use case, startups, kind of building their business, building their stack on Kubernetes. We see companies building products on top of Kubernetes. You see a lot of them here on the expo floor. That's a different type of vendor startup ecosystem. I think there's lots of opportunities there. For the big multinationals, I think one really interesting thing that hasn't really quite been done yet, is sort of treating Kubernetes as a first-class citizen as opposed to a way to commercialize and enter a new market. I think one of the default ways large technology companies tend to look at something hypergrowth like Kubernetes and TensorFlow and other projects is wrapping around it and commercializing in some way, and I think a deeper more strategic path for large companies could be to really embed Kubernetes in the core kind of crown jewel IP assets that they have. So I'll give you an example, like, for let's just take SAP, I'll just pick on SAP randomly, for no reason. This is one of the largest enterprise software companies in the world. I would encourage the co-CEOs of SAP, for example. >> John: There's only one CEO now. >> Is there one CEO now? Okay. >> John: Snabe left. It's now (drowned out by talking). >> Oh, okay, gotcha. I haven't been keeping up on the SAP... But let's just say, you know, a CEO boardroom level discussion of replatforming the entire enterprise application stack on something like Kubernetes could deliver a ton of really core meaningful benefits to their business. And I don't think like deep super strategic investments like that at that level are being made quite yet. I think at a certain point in time in the future they'll probably start to be made that way. But that's how I would like look at smart investments on the bigger scale. >> We're not seeing scale yet with Kubernetes, just the toe is in the water. >> I think we're starting to see scale, John. I think we are. >> John: What's the scale number in clusters? >> I'll give you the best example, which came up today, and actually really surprised me which I think was a super compelling example. The largest retailer in China, so essentially the Amazon of China, JD.com, is running in production for years now at 20,000 compute nodes with Kubernetes, and their largest cluster is a 5,000 node cluster. And so this is pushing the boundary of the sort of production-- >> And I think that may be the biggest one I've heard. >> Yeah, that's certainly, I mean for a disclosed user that's pretty huge. We're starting to see people actually talk publicly about this which is remarkable. And there are huge deployments out there. >> We saw Tyler Jewell come on from WSO2. He's got a new thing called Ballerina. New programming language, have you seen that? >> Joseph: I have, I have. >> Thoughts on that? What's your thoughts on that? >> You know, I think that, so I won't make any particular specific comments on Ballerina, I'm not extremely informed on it. I did play with a little bit, I don't want to give any of my opinions, but what I'd say, and I think Tyler actually mentioned this, one of the things that I believe is going to be a big deal in the coming years, is so, trying to think of Kubernetes as an implementation detail, as the kernel, do you interact directly with that? Do you learn that interface directly? Are you sort of kind of optimizing your application to be sort of natively aware of those abstractions? I think the answer to all of those questions is no, and Kubernetes is sort of delegated as a compiler target, and so frankly like directionally speaking, I think what Ballerina's sort of design is aspiring towards is the right one. Compile time abstraction for building distributed systems is probably the next logical progression. I like to think of, and I think Brendan Burns has started to talk about this over the last year or two. Everyone's writing assembly code 'cause we're swimming yaml and configuration based designs and systems. You know, sort of pseudodeclarative, but more imperative in static configurations. When in reality we shouldn't be writing these assembly artifacts. We should be delegating all of this complexity to a compiler in the same way that you know, we went from assembly to C to higher level languages. So I think over time that starts to make a lot of sense, and we're going to see a lot of innovation here probably. >> What's your take on the community formation? Obviously, it's growing, so, any observations, any insight for the folks watching what's happening in the community, patterns, trends you'd see, like, don't like. >> I think we could do a better job of reducing politics amongst the really sort of senior community leaders, particularly who have incentives behind their sort of agendas and sort of opinions, since they work for various, you know, large and small companies. >> Yeah, who horse in this race. >> Sure, and there's, whether they're perverse incentives or not, I think net the project has such a high quality genuine, like humble, focused group of people leading it that there isn't much pollution and negativity there. But I think there could be a higher standard in some cases. Since the project is so huge and there are so many very fast moving areas of evolution, there tends to be sort of a fast curve toward many cooks being in the kitchen, you know, when new things materialize and I think that could be better handled. But positive side, I think like the project is becoming incredibly diverse. I just get super excited to see Aparna from Google leading the project at Google, both on the hosted Saas offering and the Kubernetes project. People like Liz and others. And I just think it's an awesome, welcoming, super diverse community. And people should really highlight that more. 'Cause I think it's a unique asset of the project. >> Well you're involved in some deep history. I think we're going to be looking this as moment where there was once a KubeCon that was not part of the CNCF, and you know, you did the right thing, did a good thing. You could have kept it to yourself and made some good cash. >> It's definitely gotten really big, and it's way beyond me now at this point. >> Those guys did a good job with CNCF. >> They're doing phenomenal. I think vast majority of the credit, at this scale, goes to Chris Anasik and Dan Conn, and the events team at the Linux Foundation, CNCF, and obviously Kelsey and Liz and Michelle Noorali and many others. But blood, sweat, and tears. It's no small feat pulling off an event like this. You know, corralling the CFP process, coordinating speakers, setting the themes, it's a really huge job. >> And now they got to deal with all the community, licenses, Lauren your thoughts? >> Well they're consistent across Apache v2 I believe is what Dan said, so all the projects under the CNCF are consistently licensed. So I think that's great. I think they actually have it together there. You know, I do share your concerns about the politics that are going on a little bit back and forth, the high level, I tend to look back at history a little bit, and for those of us that remember JBoss and the JBoss fork, we're a little bit nervous, right? So I think that it's important to take a look at that and make sure that that doesn't happen. Also, you know, open stack and the stuff that we've talked about before with distros coming out or too many distros going to be hitting the street, and how do we keep that more narrow focused, so this can go across-- >> Yeah, I started this, I like to list rank and iterate things, and I started with this sheet of all the vendors, you know, all the Kubernetes vendors, and then Linux Foundation, or CNCF took it over, and they've got a phenomenal sort of conformance testing and sort of compliance versioning sheet, which lists all the vendors and certification status and updates and so on and I think there's 50 or 60 companies. On one hand I think that's great, because it's more innovation, lots of service providers and offerings, but there is a concern that there might be some fragmentation, but again, this is a really big area of focus, and I think it's being addressed. Yeah, I think the right ones will end up winning, right? >> Joseph: Right, for sure. >> and that's what's going to be key. >> Joseph: Healthy competition. >> Yes. >> All right final question. Let's go around the horn. We'll start with you JJ, wrapping up KubeCon 2018, your thoughts, summary, what's happened here? What will we talk about next year about what happened this week in Denmark? >> I think this week in Denmark has been a huge turning point for the growth in Europe and sort of proof that Kubernetes is on like this unstoppable inflection, growth curve. We usually see a smaller audience here in Europe, relative to the domestic event before it. And we're just seeing the numbers get bigger and bigger. I think looking back we're also going to see just the quality of end users and the end user community and more production success stories starting to become front and center, which I think is really awesome. There's lots of vendors here. But I do believe we have a huge representation of end users and companies actually sharing what they're doing pragmatically and really changing their businesses from Financial Times to Cern and physics projects, and you know, JD and other huge companies. I think that's just really awesome. That's a unique thing of the Kubernetes project. There's some hugely transformative companies doing awesome things out there. >> Lauren your thoughts, summary of the week in Denmark? >> I think it's been awesome. There's so much innovation happening here and I don't want to overuse that word 'cause I think it's kind of BS at some point, but really these companies are doing new things, and they're taking this to new levels. I think that hearing about the excitement of the folks that are coming here to actually learn about Kubernetes is phenomenal, and they're going to bring that back into their companies, and you're going to see a lot more actually coming to Europe next year. I also true multicloud would be phenomenal. I would love that if you could actually glue those platforms together, per se. That's really what I'm looking for. But also security. I think security, there needs to be a security seg. We talked to customers earlier. That's something they want to see. I think that that needs to be something that's brought to the table. >> That's awesome. My view is very simple. You know I think they've done a good job in CNCF and Linux Foundation, the team, building the ecosystem, keeping the governance and the technical and the content piece separate. I think they did a good job of showing the future state that we'd like to get to, which is true multicloud, workload portability, those things still out of reach in my opinion, but they did a great job of keeping the tight core. And to me, when I hear words like defacto standard I think of major inflection points where industries have moved big time. You think of internetworking, you think of the web, you think of these moments where that small little tweak created massive new brands and created a disruptor enabler that just created, changed the game. We saw Cisco coming out of that movement of IP with routers you're seeing 3Com come out of that world. I think that this change, this new little nuance called Kubernetes is going to be absolutely a defacto standard. I think it's definitely an inflection point and you're going to see startups come up with new ideas really fast in a new way, in a new modern global architecture, new startups, and I think people are going to be blown away. I think you're going to see fast rising growth companies. I think it's going to be an investment opportunity whether it's token economics or a venture backer private equity play. You're going to see people come out of the wood work, real smart entrepreneur. I think this is what people have been waiting for in the industry so I mean, I'm just super excited. And so thanks for coming on. >> Thank you for everything you do for the community. I think you truly extract the signal from the noise. I'm really excited to see you keep coming to the show, so it's really awesome. >> I appreciate your support, and again we're co-developing content in the open. Lauren great to host with you this week. >> Thank you, it's been awesome. >> And you got a great new venture, high five there. High five to the founder of KubeCon. This is theCUBE, not to be confused with KubeCon. And we're theCUBE, C-U-B-E. I'm John Furrier, thanks for watching. It's a wrap of day two global coverage here exclusively for KubeCon 2018, CNCF and the Linux Foundation. Thanks for watching. (techno music)

>> Announcer: Live from the Computer History Museum in Mountain View, California. It's theCUBE, covering DevNet Create 2018, brought to you by Cisco. >> Hi there and welcome to the special CUBE live broadcast here at the Computer History Museum in Mountain View, California, It's theCUBE's exclusive coverage of Cisco's DevNet Create. This is Cisco's developer ecosystem brand new, second event that they've done, and it's one and a half years in existence. This is Cisco's extension to their DevNet developer program, which is mostly Cisco developers, mostly networking, and theCUBE is here covering the future of cloud native Kubernetes, and the future of application development, as networks become more programmable. I'm John Furrier, your host, with Lauren Cooney, analyst today co-hosting with me, all day coverage. Our guest is Allan Naim, who is the product manager at Google Kubernetes engine, at Google, right down the street here. Allan, great to have you, thanks for joining us. >> Yeah, thanks for inviting me. >> So, you are the key man with the fireside chat with Susie Wee who is heading up this whole program, doing an amazing job. Google's no stranger. We all know Google at the scale level, massive scale, running infrastructure, building your own stuff, really inventing the category and then fast followers, Facebook among others, large scale. So, you guys invented Kubernetes. So that's a fact. So, tell the story of how it started because there was a moment in Google where Kubernetes, there was a debate. Do we keep it internally, open it up? And you guys have history. You've created MapReduce, you've created the data surge that we're seeing now and changing the game there. Maybe a little bit differently than how Kubernetes is handled. What's the inside story about the creation of Kubernetes and how it's evolved? >> Yeah, so Google has been working with containers for a long, long, time. It's nothing new to Google, and we wanted really to take a lot of the best practices associated with how we manage and run containers internally and share that with the community as a whole. What we found initially was the move to the cloud was very much traditionally a lift and shift and modernized move. And, there's a reason why only, I think the latest statistic I've seen is less than 10% of the applications have actually moved to the cloud. What about the other 90%? So, we wanted to bring some of the magic that Google uses internally and bring that to the world, right, so that you can modernize wherever you're running, right, for those applications that can't just move to the cloud. Why not provide a way to take advantage of some of the innovations that we've created around packaging applications up, deploying applications very seamlessly, and then eventually moving them to the cloud with less friction? And that was really behind the reason we took Kubernetes, which is really a set of best practices around how Google runs and operates containers, and made it available to the open source community. We could've kept it internally, right, and not shared it with the community, but then that really stifles innovation. Google is not about stifling innovation. We're about enabling the community to really drive innovation and build an ecosystem around it. And looking back now, it was a tremendous move. >> Yeah, and you know what, the leadership I remember at that time, and I wanted to get that out there. Thank you for sharing that. Craig McLuckie, Brendan Burns, Joe Beda, those guys and the team around them, it was kind of a small team, held the line on that. And the conversation was, this needs to happen in an open way mainly because you saw, though, how to manage your workloads internally and wanted to bring it to the masses. So, real props to the original team, a really good call, and again, it worked out great. >> Yes. >> So, okay, today. Where are we today? Because now you go back at the creation of Kubernetes, you guys open it up, still contributed and nurtured it, and now it becomes part of the bigger part of the open source community. You have now new innovations. What is the update from your standpoint where Kubernetes is today? Okay, it's well know that the containers is now standard and standard. Now the business model container hasn't materialized. That's okay. The technical architecture is very solid. Kubernetes has become the favorite child in the architecture because of the benefits. What's the update? What's Kubernetes doing today that's compelling? What's the update? >> Yeah, so just as you said, containers are mainstream now. Kubernetes is on fire. We see a world today where Kubernetes is literally running everywhere, right, from Google Cloud to other clouds to partnerships that we have with the likes of Cisco. You now have these clusters that are popping up in heterogeneous environments. So, we've enabled developers now to really build services very efficiently and update those services in a consistent manner regardless of where those services are running. Now, as you build more and more clusters and expose more and more services, the day two experience starts coming in, right? How do I manage this environment? How do I manage my services? How do I find out what these services are actually doing, which services are talking to each other? How do I do more of the networking aspect around traffic management? And this is where I see a lot of the investments happening right now in the open source world with projects like Istio, which are fairly new, but are taking a lot of the goodness that Kubernetes is bringing and applying more of an operations mindset around networking. >> And what problem is that solving? Can you be specific? Because I like this day two experience. I mean, day three will be like, oh my God. How do you manage it beyond that, but, what is the problem that's being solved? Is it more industrial strength, is it more tolerance? Is it securities or all the above? What's the main problem? >> It's security, it's when you're running services in heterogeneous environments, there is no consistent security model, right? Istio helped solved some of that. It's service discovery. When services are running, again, in environments where you have different mechanisms for storing services, how do you discover these services? Now, how do you route traffic to the right service? How do you do canary deployments where perhaps I'd like to trickle certain load onto a new version and eventually move all my work into the new version that I've deployed? So, canary testing. Running services in geographic locations and using networking algorithms to route my requests to the closest location. Those are all really hard challenges that you need to solve, and technologies like Istio really make it possible for developers to get those benefits without having to write a single line of code, right? So, you leverage the API to get all these benefits that I just talked about. >> I want to get you in for a minute to talk about that if we can. Talk about Google cloud right now vis a vis the momentum because a lot's changed with Google just in the past couple of years. A lot of people on board, new hires, industry veterans, leaders. We've heard Lou Tucker from Cisco say at CubeCon that Istio is probably the biggest thing he's seen in years in terms of its implementation capability to impact the valued creation of application developers and also in creating efficiencies in networks. How is the Google team right now doing? Give an update, because you guys are now in the center of it and I've called you guys, the real competitor to Amazon, because I consider you and amazon probably the coolest cloud and most relevant clouds vis a vis what clients want to do in a modern era. Not so much retrofitting legacy cloud to make it kind of retrofit, but really doing ground zero cutting edge cloud stuff. What's the update from Google Cloud? What are you guys most proud of? What's the things that you want to highlight that are notable? >> So, Google Cloud's been growing at a tremendous rate. It's just mind-boggling how fast customer adoption has been. What we've seen is, the adoption has spanned all the way from startup to small, medium-sized businesses, extending into the Fortune 100s regardless of industry. And what we hear from customers is they like the clean APIs that Google provides. They like our compute infrastructure from a resiliency standpoint, the transparency that we provide in terms of enabling customers in running their workloads on Google Cloud. We've made a lot of investments in Google Cloud and we continue to make these investments. Now, on the cloud native and container fronts, what we're doing and what we're focusing on is really a differentiated model where we are working with customers to enable them to modernize in place and move to the cloud at their own pace versus having to lift and shift an application to take advantage of modernization and APIs in the cloud. That's really a differentiating story that we're bringing to the table. Along with that, we continue to invest in storage, in optimizing our networking, in setting up more and more points of presence around the world. We added, I believe, over 12 zones last year around the world. So, the growth rate has just been phenomenal. On the Kubernetes side, it's all about value, right? It's all about differentiated value as well. Google has been operating a managed Kubernetes service now for over two years. Building and providing a managed service is hard, right? We have the expertise to do that. We feel that Google Cloud is the best environment on the planet for running containers. And through this expertise, we'll continue to invest to bring our services and make it a first class experience to run managed scale containers as well. >> So, would it be safe to say that you guys are focused on differentiating and not trying to be the whole world, everything to everybody, to really kind of narrow the focus? >> Well, there are table stakes that you need to address, especially around storage and networking, and we feel we've gotten there, right? Now, for a customer that's picking a cloud, whether it's Google or any other cloud, we've addressed those table stakes. But on the cloud native side of the house, when building containerized applications, we feel that we have a differentiated offering that really no other cloud on the planet can deliver on. >> That's awesome. Let's talk about, and my last question is mush more about developers' relationship to the new architecture. We'll call this the new architecture. >> Yeah. >> You've got Kubernetes which has done some great innovate work, containers continue to be a great resource aspect of architecture, and storage infrastructure becoming more programmable like what Cisco's offering. Great stuff. App developers. I just want to write code. So, you've got some developers. How does a developer, in your opinion, Google's opinion, yours and Google's opinion, how do they determine their relationship to the network or the new architecture? You've got some guys who just want to write apps. So, I don't want to do any kind of speeds and feeds. Some guys want to get down and dirty and wire up some services when you get in the middle layer, and some might want to get down low in the stack. How does a developer kind of peg their orientation to different parts of the cloud architecture? >> So, when you really think about it, Kubernetes is a logical layer that sits on top of infrastructure that makes it possible to take an application that runs a certain way in one location to run consistently in other locations. So, for application developers that just want to write code, we've got a clean set of APIs that they can take advantage of to spin up cluster resources, deploy their applications. We've been heavily focused as well on not just creating an amazing story for stateless applications, but stateful applications as well. So, being able to orchestrate, you choreograph your application deployment. Now, for developers that want to get their hands dirty, the way we've designed Kubernetes is very much an extensible model. So, the Kubernetes APIs can be extended and functionality can actually be over written to tailor the experience. A developer may want to plug in a different type of controller, for example, versus the standard Kubernetes controller. So, we enabled that, think of it as a peel the onion approach, so that we can meet the developer where they are and give them the tools required for them to actually be productive in their companies or in the community. >> Awesome. Right, and you guys have a deal with Cisco, or relationship with Cisco, or else you're here, at the DevNet Create event, which is about cloud native, not so much about being kind of Cisco or DevNet, the classic developer program. On stage you talked about Istio. Is that the key to the partnership with Cisco? What specifically is your relationship to Cisco? >> Yeah, it's a great question. So, with Cisco, we've been hearing from customers a lot that getting Kubernetes up and running on premise is really hard. We've also been hearing a lot from customers that they want support. So, we got together with Cisco to provide a hybrid offering that tailors customers that want to start their journey to cloud native on prem. So, Cisco basically provides a mechanism, right, for customers to actually run Kubernetes on prem with a single support model for all their needs, which is great for Google because this is something that Cisco-- >> They know a lot about that. >> Absolutely. Now, for customers that want to start building in the cloud and connecting to the cloud, but you need secure performance networking. How do you do that, right? Well, Cisco is an innovator in networking and security. Google is an innovator in cloud and open source technology and cloud native technology. So, we bring these two things together to give really developers and sys admins a world where they can collaborate and have an API-driven approach to running workloads that span a hybrid estate. >> John: And it's great for you guys too. You open up your market to the enterprise. >> Yeah, I would say that also it really gives an opportunity for network engineers and developers, and I think you talked about clusters ops and Arkino and new types of app ops that you're bringing to the table-- >> Yep, yes. >> And what kind of roles do you see these people playing as you grow that ecosystem? >> Exactly. It's not just about the technology, but it's the culture within the company that oftentimes really drives, it's a hard obstacle to bypass. For customers that I talk to, a lot of times they tell me, look, we've settled. We want to go with Kubernetes, but what about the internal culture? How do we build our teams around Kubernetes? How do we scale our services in such a way where we have specialization of service?kino And I talked about Narkino, the whole notion of separation of concerns where we introduce this new notion in terms of how Google does things of an application ops team that's typically small in size, but their role starts where the developer role ends, and basically, they're responsible for taking an application from a developer and deploying it out into a environment. Then you have a cluster ops role team that's focused on the underlying infrastructure and maintains all the various cluster APIs, the Kubernetes environment. So, think of them as shared services that are very much tailored to enabling developers to do what they do best and build great applications and push changes in production very quickly. >> Well, thanks for coming out to theCUBE. I know you've got another hard stop. You're got another panel. Real quick, I'll give you the final word. What's the one thing people should know about Google Cloud that they may not know about or gets buried in the noise out in the marketplace? >> Yeah. Google Cloud is the most innovative cloud out there on the market. We have points of presence in literally every region around the world. Our APIs are some of the cleanest out there of any cloud, as well as the Kubernetes experience running in Google has been something that we've been invested in for over two years and it's actually a highly optimized experience for developers that want to run their containerized application and very differentiated. And 100% upstream compatible with Kubernetes open source. >> That's great stuff. I got to tell you, just Google team, we covered all the cloud players from day one. There's no shortcut. You've got to put the work in, whether it's public sector or getting the building blocks in there. You guys do a great job. Congratulations. >> Thank you. >> Kubernetes is worth noting. theCUBE covering all the action, and the story here is Kubernetes, Google's creation, which is now open standard for all, 100% upstream compatible here at the Cisco's DevNet Create event. Back with more live coverage. I'm John Furrier with Lauren cooney after this short break. (upbeat music) [Announcer] In center.

from Austin Texas it's the cube covering cube con and cloud native con 2017 brought to you by Red Hat the Lenox foundations and the cubes ecosystem partners hello everyone welcome to the cube live in Austin Texas for exclusive coverage of cloud native conference and cube con cube con with the linux foundation on john fourier co-founder silicon angle media tube Minutemen with ricky bond and also covering the developer community we just came off Amazon reinvent last week we're now in Austin Texas for a continuation of the Builder theme around this new generation of developers exclusive coverage of cloud native con and cube con or cube cons to cube con like kubernetes john byrne not to be confused with the cube of course when 2018 we're gonna do cube con right John yeah so cube con is coming to check for local listings around an area near you will be will be there stew what a great event I love this events one of my favorite events as you know personally for the cube audience out there and who know us I've been following us we've been growing up with this community we've been covering the Linux Foundation from the beginning if you go back to our roots around 2010 we've always been on the next wave whether it was big date of the converge infrastructure on the enterprise and then cloud the cube is always on the wave and then wave and we call that we were there when kubernetes was formed we were there with the principles JJ when and his team cuz Maddock with blue Tucker kind of brain so me hey we should do kubernetes and we said then kubernetes would be huge it would be the orchestration that would be the battleground in what we were at the time calling the middleware of the cloud turns out that was true that is happening huge change in the ecosystem as containerization with docker originally starting it and then the evolution of how software developers are voting with their workloads they're voting with their code and no better place than the Linux Foundation to your analysis obviously we're super excited but there's some dynamics going on there's a class of venture backed companies that I won't say are groping for a strategic position are certainly investing in open source but brings up the questions of the business model where's the value being created what is the right strategy do I do services do I have a different approach there's a lot of different opinions and if the customers choose wrong they could be on the wrong side of history as this massive wave of innovation with AI machine learning is impacting infrastructure and DevOps it's awesome we heard Netflix on stage let's do what's your take what's going on here cloud native cons yeah so so John I love you know Dan who were you know runs the CN CF gets out on stage and he says you know it's exciting time for boring infrastructure maybe maybe too exciting I even said you know we've been watching this wave of you know containerization and kubernetes and this whole CN CF ecosystem has really taken you know that container piece and exploded beyond this really talking about how I build for these cloud native environments you know there's 14 projects here kubernetes is the one that kicked it off but so many pieces of what's happening here john AWS last week phenomenal like 45,000 people a lot of the real builders the ones you know heavily involved in projects or like ah I actually might skip AWS come to come to coop con this coop con this is where you know so many people we've seen you know founders of companies working on so many projects you know large community you know great community focus I know you like Netflix up there talking about culture big diversity I think what was it 130 scholarships for people of diversity there so really phenomenal stuff you know this is where really that multi-cloud world is being built yeah and good points too because that's really the elephant in the room which is the prophets and the monetization of developer communities is not the primary but it's a big driver and how people are behaving and Amazon reinvent in this world are parallel universes you know it's interesting you don't see a lot of reinvent hoodies I wore mine last night got a couple dirty looks but this is you see a lot of Google you see a lot of Microsoft John John John we have Adrian Cockcroft was in the keynote this morning everybody's saying we're braising the databases here you don't think that's the case I think everyone does embrace it isn't number one isn't really no second place they're far back as I said at reinvent I still stand by that but you got big players okay dan cohen basically said on stage okay it's my projects products and profits and they're putting profits actually in the narrative because they're not shying away from soup but it's not a pay-to-play kind of ecosystem here it's like saying look at the visibility of the cloud has shine the light on the fact that there is an opportunity to create value of which value then can be translated to monetization and developers like to get paid no one likes that do things totally for free that is the scoreboard of value it's not just about chasing the dollar and I think I like how the CN CF is putting out the prophets saying look at this real value here in businesses is real value in products that come from these projects this is a new era and open source I think that's legit again pay-to-play is a completely different animal yeah vendors come in control the standards pay pay pay not anymore Brendan burns told me last year Microsoft no pay to play Microsoft's got a big platform they're gonna come in and make things happen ok so John the money thing is a big question I have coming into this week dan talked up on stage there's certified service providers for kubernetes and there's certified kubernetes partners 42 certified kubernetes partners for the most part kubernetes has been commoditized today that certification doesn't mean that hundred-percent everything works but it definitely over a you know short period of time it will be means that I if I choose any platform that uses kubernetes that certified I can move from one to the other it doesn't mean that I'm actually going to make money selling kubernetes it's that that's part of the platform or services that arm offering and it is an enabler and you know that's what's a little different you think about you know John we try for years OpenStack thought we were gonna make money on it how we're gonna make money even go back to Linux you know it's what can be built using this set of tools so people have said this is really rebuilding the analytics for the cloud environment but money is kind of its derivative off of it it's an enabler to are there great software it brilliants dude this is the bottom line here it's the tale of two stories in the industry okay this in the backdrop is this and if prices are an IT specifically in development teams platforms are shifting big time the old is an old guard as Andy Jackson said the invest in a new guard the dynamics are containerization drove megatrend number one that turbocharged the cloud infrastructure and gave developers some freedom micro-services then take it to another level what it's actually done has changed at two theaters in the industry theater one is the vendors that are getting funded that participants in open source work trying to create value and then what I would call the rest of the market there is an onboarding a tsunami of new developers coming in I'm seeing in the in theater one all the people that we know in the industry and then I'm seeing new faces these are people who are going to the light the light is the monetization and that's the value creation so you seeing people here for the first time you're seeing developers who have a clear line of sight that this community creates value so that's two dynamics so that the companies that got a hundred million dollars in funding from venture capitals they're trying to figure out can they take advantage of that wave of new developers there's been an in migration into cloud native of new developers and these are the ones they're going to be creating the value the creativity the solutions and certainly the cultural impact from those solutions will be great I see a great opportunity if people just don't get scared and just hold the line keep your hitting value it'll figure itself out so the evolution is natural and that is something I'm interesting to see okay and John the thing I'm looking for this week first of all when we talked about containers we talked about this whole cloud native environment that boring infrastructure stuff it still matters networking has matured a little bit there's the CNI initiative the cloud native I'm sorry container networking interface which is approaching one auto they're getting feedback here second one is storage most of the these solutions we really started talking about stateless environments state absolutely has to be a piece of this how do we fit you know you know data AI ml all these things data is critically critically important so that needs to be there and then the new technology that you know we spent a lot of time talking about at AWS that was serverless and there's actually like a half day track here at this show talking about how all of these solutions how serverless fits into them there was a question does serverless replace the because I don't need to think about it really a lot of the same tooling a lot of these usage will fit into those server lists frameworks so it's not in either/or but really more of an an environment but definitely something that we expect to hear more of this we've done we've got a phenomenal lineup I'm super excited did you know some of these builders that we've got you know big players we've got startups we've got authors we've got a good diverse audience coming on the cube so and you know I know near and dear to your heart you know lots of developer talk a lot of their over talks do this is a fun time the commoditization of kubernetes is actually a good thing in my mind I think there will be a lot of value to be created and this really is about multi cloud you mentioned all three of the major clouds and now Maurice are all a bob on stage just in China you got a lot more growth you're seeing that kubernetes really is an opportunity for Google and Microsoft and the rest of the community to run as fast as they can to create services so that customers can have a choice choice is the new black that's what's going on and multi-cloud not yet here but certainly on the horizon and if Google and Azure do not establish a mike-mike multi cloud environment Amazon could run away with it that's my that's my tag that's my visibility on it the bottom line is whoever can creates the value so what I'm gonna look for is the impact of the continued kubernetes kinda monetization and the new formations do the new relationships the existing players like red hat are going to continue to kick ass you're gonna start to see new players come in you can expect to see new partnerships because the stack is being developed very fast smooth announcements for me theists flew and deke container D Windows support coming with 1.9 kubernetes what's happening is they're running as fast as they can they're pedaling as fast as they can because if they do not they will be blown away that's the cube coverage here kicking off day one I'm John Purdue minimun exciting times here at cloud native and cube on back after this short break

>> Announcer: Live from Midtown Manhattan, it's theCUBE, covering Big Data, New York City, 2017. Brought to you by SiliconANGLE Media and its ecosystem sponsors. >> Okay, welcome back, everyone. Live in Manhattan, this is theCUBE's coverage of our fifth year doing Big Data, NYC; eighth year covering Hadoop World, which is now evolved into Strata Data which is right around the corner. We're doing that in conjunction with that event. This is, again, where we have the thought leaders, we have the experts, we have the entrepreneurs and CEOs come in, of course. The who's who in tech. And my next two guests, is Jagane Sundar, CUBE alumni, who was on yesterday. CTO of WANdisco, one of the hottest companies, most valuable companies in the space for their unique IP, and not a lot of people know what they're doing. So congratulations on that. But you're here with one of your partners, a company I've heard of, called Microsoft, also doing extremely well with Azure Cloud. We've got Pranav Rastogi, who's the program manager of Microsoft Cloud Azure. You guys have an event going on as well at Microsoft Ignite which has been creating a lot of buzz this year again. As usual, they have a good show, but this year the Cloud certainly has taken front and center. Welcome to theCUBE, and good to see you again. >> Thank you. >> Thank you. >> Alright, so talk about the partnership. You guys, Jagane deals with all the Cloud guys. You're here with Microsoft. What's going on with Microsoft? Obviously they've been, if you look at the stock price. From 20-something to a complete changeover of the leadership of Satya Nadella. The company has mobilized. The Cloud has got traction, putting a dent in the universe. Certainly, Amazon feels a little bit of pain there. But, in general, a lot more work to do. What are you guys doing together? Share the relationship. >> So, we just announced a product that's a one-click deployment in the Microsoft Azure Cloud, off WANdisco's Fusion Replication technology. So, if you got some data assets, Hadoop or Cloud object stores on-premise and you want to create a hybrid or a Cloud environment with Azure and Picture, ours is the only way of doing Active/Active. >> Active/Active. And there is some stuff out there that's looking like Active/Active. DataPlane by Hortonworks. But it's fully not Active/Active. We talked a little bit about that yesterday. >> Jagane: Yes. >> Microsoft, you guys, what's interesting about these guys besides the Active/Active? It's a unique thing. It's an ingredient for you guys. >> Yes, the interesting thing for us is, the biggest problem that we think customers have for big data perspective is, if you look at the landscape of the ecosystem in terms of open source projects that are available it's very hard to a: figure out How do I use this software?, b: How do I install it? And, so what we have done is created an experience in Azure HDInsight where you can discover these applications, within the context of your cluster and you can install these applications by one-click install. Which installs the application, configures it, and then you're good to go. We think that this is going to sort of increase the productivity of users trying to get sense out of big data. The key challenges we think customers have today is setting up some sort of hybrid environment between how do you connect your on premise data to move it to the Cloud, and there are different use cases that you can have you can move parts of the data and you can do experiment easily in the Cloud. So what we've done is, we've enabled WANdisco as an application on our HDInsight application platform, where customers can install it using a single-click deploy connected with the data that's sitting on-prem, use the Active/Active feature to have both these environments running simultaneously and they're in sync. >> So one benefits the one-click thing, that's on your side, right? You guys are enabling that. So, okay, I get that. That's totally cool. We'll get to that in a second. I want to kind of drill down on that. But, what's the benefit to the customers, that you guys are having? So, I'm a customer, I one-click, I want some WANdisco Active/Active. Why am I doing it? What does the Cloud change? How does your Cloud change from that experience? >> One example that you can think about is going to change is in an on-premise environment you have a cluster running, but you're kind of limited on what you can do with the cluster, because you've already setup the number of nodes and the workloads your running is fairly finite, but what's happening in reality and today is, lots of users, especially in the machine learning space, and AI space, and the analytic space are using a lot of open source libraries and technologies and they're using it on top of Hadoop, and they're using it on top of Spark. However, in experimenting with these technologies is hard on-prem because it's a locked environment. So we believe, with the Cloud, especially with it offering WANdisco and HDInsight, once you move the data you can start spinning up clusters, you can start installing more open source libraries, experiment, and you can shut down the clusters when you're done. So it's going to increase your efficiency, it's going to allow you to experiment faster, and it's going to reduce for cost as well, because you don't have to have the cluster running all the time and once you are done with your experimentation, then you can decide which way do you want to go. So, it's going to remove the-- >> Jagane, what's your experience with Azure? A lot of people have been, some people have been critical, and rightfully so. You guys are moving as fast you can. You can only go as fast you can, but the success of the Cloud has been phenomenal. You guys have done a great job with the Cloud. Got to give you props on that. Your customers are benefiting, or Microsoft's customers are benefiting. How's the relationship? Are you getting more customers through these guys? Are you bringing customers from on-prem to Cloud? How's the customer flow going? >> Almost all of our customers who have on-prem instances of Hadoop are considering Cloud in one form or the other. Different Clouds have different strengths, as they've found-- >> Interviewer: And different technologies. >> Indeed. And Azure's strengths appear to be the HDInsight piece of it and as Pranam just mentioned, the cool thing is, you can replicate into the Cloud, start up a 50 node Spark cluster today to run a query, that may return results to you really fast. Now, remember this is data that you can write to both in the Cloud and on-premise. It's kept consistent by our technology, or tomorrow you may find that somebody tells you, Hive with the new Tez enhancements is faster, sure, spin up a hundred node Hive cluster in the Cloud, HDInsight supports that really well. You're getting consistent data and your queries will respond much faster than your on-premise. >> We've had Oliver Chu on, before with Hortonworks obviously they're partnering there. HDInsight's been getting a lot of traction lately. Where's that going? We've seen some good buzz on that. Good people talking about it. What's the latest update on your end? >> HDInsight is doing really good. The customers love the ease of creating a cluster using just a few clicks and the benefits that customers get, clusters are optimized for certain scenarios. So if you're doing data science, you can create a Spark cluster, install open source libraries. We have Microsoft R Server running on Spark, which is a unique offering to Microsoft, which lots of customers have appreciated. You also have streaming scenarios that you can do using open source technologies, like we have Apache Kafka running on a stack, which is becoming very popular from an ingestion perspective. Folks have been-- >> Has the Kupernetes craze come down to your group yet? Has it trickled down? It seems to be going crazy. You hired an amazing person from Google, Brendan Burns, we've interviewed before. He's part of the original Kubernetes spec he now works for Microsoft. What's the buzz on the Kubernetes container world there? >> In general, Microsoft Azure has seen great benefits out of it. We are seeing lots of traction in that space. From my role in particular, I focus more on the HDInsight big data space, which is kind of outside of what we do with Kubernetes' work. >> And your relationship is going strong with WANdisco? >> Pranav: Yes. >> Right. >> We just launched this offering just about yesterday is what we announced and we're looking forward to getting customers on to the stack. >> That's awesome. What's your take on the industry right now? Obviously, the partnerships are becoming clearer as people can see there's (mumbles). You're starting to see the notion of infrastructure and services are changing. More and more people want services and then you got the classic infrastructure which looks like it's going to be hybrid. That's pretty clear, we see that. Services versus infrastructure, how should customers think about how they architect their environments? So they can take advantage of the Active/Active and also have a robust, clean, not a lot of re-skilling going on, but more of a good organization from a personnel standpoint, but yet get to a hybrid architecture? >> So, it depends, the Cloud gives you lots of options to meet the customers where they are. Different customers have different kinds of requirements. Customers who have specialized, some of their applications will probably want to go more of an infrastructure route, but customers also love to have some of the past benefits where, you know, I have a service running where I don't have to worry about the infrastructure, how dispatching happen, how does OS updates happen, how does maintenance happen. They want to sort of rely on the Microsoft Azure Cloud provider to take care of it. So that they can focus on their application specific logic, or business specific logic, or analytical workloads, and worry about optimizing those parts of the application because that is their core-- >> It's been great.I want to get your thoughts real quick. Share some color. What's going on inside Microsoft? Obviously, open source has become a really big part of the culture, even just at Ignite. More Linux news is coming. You guys have been involved in Linux. Obviously, open source with Azure, ton of stuff, I know is built in the Microsoft Cloud on open source. You're contributing now as to Kubernetes, as I mentioned earlier. Seems to be a good cultural shift at Microsoft. What's the vibe on the open source internally at Microsoft? Can you share, just some anecdotal insight into what's the vibe like inside, around open source? >> The vibe has increased quite a lot around open source. You rightly mentioned, just recently we've announced a SQL server on Linux as well, at the Ignite conference. You can also deploy a SQL server on a docker container, which is quite revolutionary if you think about how forward we have come. Open source is so pervasive it's almost used in a lot of these projects. Microsoft employees are contributing back to open source projects in terms of, bug fixes, feature requests, or documentation updates. It's a very, very active community and by and large I think customers are benefiting a lot, because there are so many folks working together on open source projects and making them successful and especially around the Azure stack, we also ensure that you can run these open source workloads lively in the Cloud. From an enterprise perspective, you get the best of both worlds. You get the latest innovations happening in open source, plus the reliability of the managed platform that Azure provides at an enterprise scale. >> So again, obviously Microsoft partnership is huge, all the Clouds as well. Where do you want to take the relationship with Microsoft? What happens next? You guys are just going to continue to do business, you're like expecting the one-click's nice, I have some questions on that. What happens next? >> So, I see our partnership becoming deeper. We see the value that HDInsight brings to the ecosystem and all of that value is captured by the data. At the end of the day, if you have stale data, if you have data that you can't rely on the applications are useless. So we see ourselves getting more and more deeply embedded in the system. We see of ourselves as an essential part of the data strategy for Azure. >> Yeah, we see continuous integration as a development concept, continuous analytics as a term, that's being kicked around. We were talking yesterday about, here in theCUBE, real time, I want some data real time and IT goes back, "Here it is, it's real time!" No, but the data's three weeks old. I mean, real time (laughs) is a word that doesn't mean I got to see it really fast, low latency response. Well, that's not the data I want. I meant the data in real time, not you giving me a real time query. So again, this brings up a mind shift in terms of the new way to do business in the Cloud and hybrid. It's changing the game. As customers scratch their heads and try to figure out how to make their organizations more DevOps oriented, what do you guys see for advice for those managers, who are really getting behind it, really want to make change, who kind of have to herd the cats a little bit, and maybe break out security and put it in it's own group? Or you come and say, okay IT guys we're going to change into our operating model, even on-prem, we'll use some burst in to the Cloud, Azure's got 365 on there, lot of coolness developing. What's the advice for the mindset of the change agents out there that are going to do the transformation? >> My advice would be, if you've done the same thing by hand over two times, it's time you automated it, but-- >> Interviewer: Two times?! >> Two times. >> No three rule? Three strikes you're out? >> You're saying two, contrarian. >> That's a careful statement. Because, if you try automating something that you've never actually tried by hand, that's a disaster as well. A couple times, so you know how it's supposed to work. >> Interviewer: Get a good groove on it. >> Right, then you optimize, you automate, and then you turn the knobs. So, you try a hundred node cluster, maybe that's going to be faster. Maybe after a certain point, you don't get any improvements, so you know how to-- >> So take some baby steps, and one easy way to do it is to automate something that you've done. >> Jagane: Yes, exactly. >> That's almost risk-free, relatively speaking. Thoughts, advice to change agents out there. This is your industry hat on. You can take your Microsoft hat off. >> Baby steps. So you start small, you get familiar with the environment and your toolsets are provided so that you get a consistent experience on what you were doing on-prem and sort of in a hybrid space. And the whole idea is as you get more comfortable the benefits of the Cloud far outweigh any sort of cultural changes that need to happen-- >> Guys, thanks for coming on theCUBE, really appreciate it. Thoughts on the Big Data NYC this week? What do you think? >> I think it's a conference that has a lot of Cloud hanging over it and people are scratching their heads. Including vendors, customers, everybody scratching their head, but there is a lot of Cloud in this conference, although this is not a Cloud conference. >> Yeah, they're trying to make it an AI conference. A lot of AI watching certainly we're seeing that everywhere. But again, nothing wrong hyping up AI. It's good for society. It really is cool, but still, that's talking about baby steps, AI is still not there. It seems like, AI from when I got my CS degree in the 80's, not a lot innovation, well machine learning is getting better, but, a lot more way to go on AI. Don't you think? >> Yes, you know a few of the announcements we've made in this week is all about making it easier for developers to get started with AI and machine learning and our whole hope is with these investments that we've done and Azure machine learning improvements and the companion app and the workbench, allows you to get started very easily with AI and machine learning models and you can apply and build these models, do a CICD process and deploy these models and be more effective in the space. >> Yeah and also the tooling market has kind of gotten out of control. We were just joking the other day, that there's this tool shed mindset where everything is in the tool shed and people bought a hammer and turned it into a lawnmower. So it's like, you got to be careful which tools you have. Think about a platform. Think holistically, but if you take the baby steps and implement it, certainly it's there. My personal opinion, I think the Cloud is the equalizer. Cloud can bring compute power that changes what a tool was built for. Even, go back six years, the tools that were out there even six years ago are completely changed by the impact of unlimited, potentially unlimited capacity horsepower. So, okay that resets a little bit. You agree? >> I do. I totally agree. >> Who wins, who loses on the reset? >> The Cloud is an equalizer, but there is a mindset shift that goes with that those who can adapt to the mindset shift, will win. Those who can not and are still clinging to their old practices will have a hard time. >> Yeah, it's exciting. If you're still reinventing Hadoop from 2011 then, probably not good shape right now. >> Jagane: Not a good place to be. >> Using Hadoop is great for Bash, but you can't make that be a lawnmower. That's my opinion. Okay, thanks for coming on. I appreciate it (laughs) You're smiling, you got something that you, no? >> Pranav: (laughs) Thank you so much for that comment. >> Yeah, tool sheds are out there, be careful. Guys do your job. Congratulations on your partnership, appreciate it. This is theCUBE, live in New York. More after this short break. We'll be right back.

>> Live from San Francisco. It's theCUBE. Covering DevNet Create 2017. Brought to you by Cisco. >> Welcome back everyone. We're here live in San Francisco for theCUBE's exclusive two days of coverage for Cisco Systems' inaugural event called DevNet Create extension. DevNet their classic developer program, for the Cisco install base of network routers. Now going to the cloud, native, going to the developer where dev-ops and the enterprise are connecting. I'm John Furrier, my cohost Peter Burris. Next is Dan Kohn, who is the Executive Director of the Cloud Native Compute Foundation, CNCF. Formerly known as Kubecon. Which is the event, Kubecon.io. Dan, great to see you. Executive Director, how's business, is going good? >> Fantastic! (John laughs) Yeah, six months ago we chatted at our last event in Seattle. And it's just amazing to see the progress since then. Projects members. >> It's been a whirlwind. Even I can't keep track. You guys are announcing all these new projects. What's the current count of projects that you guys have under the Cloud Native Compute Foundation? >> So we're up to 10. I should definitely start with the fact that Kubernetes is the anchor 10 in our original project. In a lot of ways, foundation was setup around that. And that project is just continuing to do incredibly well. Where it's one of the highest velocity projects in the history of open source. In terms of number of authors, number of commits, poll requests, issues. But now we have a constellation of other projects that are in support of that one. It can be used in a lot of different ways. >> John: Yeah. >> That we've been adding in. >> We had Craig McLuckie on earlier. Now he's with Heptio. Again, when he was doing that work, at Google, back in the days with what's his name from Microsoft now. >> Peter: Brendan Burns. >> Brendan Burns, yeah. >> Now it's an interesting question, where you say, oh, wait a minute, the three sort of key people behind Kubernetes, Craig McLuckie, Joe Beda, who's his co-founder at Heptio, then Brendan Burns, they all left Google. Is this a bad sign for the project and the technology? >> John: No, I don't think so. >> And we would say it's a spectacularly good sign. Now, if they had left and said, ah you know, containers, I'm going to do virtual machines. But in fact they said, there's such an enormous market for this. And to have Microsoft and Azure step in and say, we really want to invest in this space and we want to bring on one of the co-founders, Brendan. And for the other two co-founders, say, hey Google is making a huge investment. But we also think there's an opportunity for independent venture funded startup. >> Craig is completely passionate about this because there is an interoperability ethos that's always been around the open web. >> Dan: Umhmm. >> And certainly open source has the same ethos. Cloud Native brings an interesting thing, and it's clear now to people that there's not going to be one cloud winning them all. >> It's a multi-could world. >> Dan: Right. >> How is the Cloud Native Foundation floating in the open source world? Is it gravitating towards more infrastructure, more edge, software edge? Are you guys kind of in the middle? Are you guys the glue layer? How do you view that? >> Sure. So one way of looking at what we're doing is, helping to build a stack of software. That allows you to run your applications either on bare metal in your own data center or on any of the public clouds. Or hybrid solution where you're mixing back and forth. But the key idea is that all the core parts of that are open source. They're supported by multiple different vendors. And what that means is, you get to avoid lock-it. So today, Amazon web services has some of the most extraordinary engineering. They have all these great services that make it very easy to go onboard. But if you build your whole architecture around that, then you're stuck with AWS forever. And when time goes up, time to renegotiate your contract in a year or two, you're back again and don't have a lot of leverage. Where we think AWS is fantastic platform to run Kubernetes, to run our other projects on top of. But we don't think you want to lock-in to those services to such a degree. >> Okay, when I'm on, first of all, pretend I'm Amazon, I'm a competitive strategist, lock-in, I got to get you locked-in. I'm just going to run Kubernetes on Amazon. Why don't I just do that? >> We think that's a great solution. >> John: You do? >> Heptio and lots other folks make it very easy to run Kubernetes on Amazon. But we also think you should at least look at Kubernetes on Bluemix, on Google, on Azure. And know that in the future when you're negotiation comes up, even if you never leave, you at least threaten to leave. That you're not locked into that one vendor forever. >> So if you think about how the cloud industry structure is starting to layout, you knew we were going to have IAAS. >> Dan: Umhmm. >> SAS has been around for quite sometime. >> Dan: Right. >> The big question is what happens with that platform as a service. >> The developer world. >> Dan: Yeah. Some people think it's going to end up in the IAS element. >> Dan: Umhmm. Some people end up in the SAS. If it ends up in the IAS, you got the lock-in. Do you see a world going forward where developers have their own place, where they go and build and create software independent of either target but then add it to the various platforms. Is that a direction that you think this is all going to end up in? >> I do. Our view is that Heroku, which really invented this platform as a service concept or popularized it. You do, get push Heroku and magically your application's up. And then Cloud Foundry which came along and created a open source version of that. Those were two building blocks. But the Cloud Native essentially taking that scenario and saying, hey, that continuous integration, continuous deployment pipeline, that ability to deploy your software dozens of times per day, that's an absolute table ante for being a modern company. Not just a software company but arguably every company today needs to be doing software development like that. And then Cloud Native is a whole set of infrastructure around that to allow you to, not just have that environment in development but also to push it into production. >> So compare and contrast, based on your vision >> Dan: Umhmm. >> of how things are going to play out. A developer spends her time today doing this, and in three years, she's going to spend her time doing that. Kind of give us a sense of how >> Dan: Sure. >> you think it's going to play out. >> The simplest way to say it is that, Docker came along a few years ago, and was incredibly transformative technology for software development. It solved this really basic problem that, you hire a new employee and does it take her an entire day or entire week to get her environment together. Or can she just copy over the document container and be ready to go. And so I would argue it had the fastest uptake of any developer technology in history. But now when you have all those pieces running, okay, that's great in development, how do you get it in production? And my goal is that in a few years, hopefully much sooner, that those developers that are getting the container, they're getting the different pieces of microservices working. And then it's this tiny little YAML file that just says, here's the requirements for my application, here's what kind of redundancy it needs, what is backend databases, other sorts of things. And they're deploying it up. For most developers they can get out of that business of dev-ops. Of having to worry about all those issues. Your dev-ops team can be so much more efficient cuz Kubernetes and the related platform really enables that. >> I got to ask you, I just Tweeted cuz I had, make sure I captured it. I'm blown away by your success on the sponsorship participation. And usually it's a sign of opportunity. Because there's money making to be made, having the big vendors in there. But the growth of Kubernetes as you mentioned, all the success, we're well aware of that. But you got a lot going on. You're like got the tiger by the tail, your hair's blown back, you're running as hard as you can. Why are you guys successful? What is your gut? As executive director, you got to have the 20 mile stare but you also implement the here and now. >> Dan: Sure. >> How are you rationalizing the success? >> The most important point is, there's not some sort of magic formula, that CNCF has done or the Linux Foundation. And we're just so much better promoting or marketing it. At the end of the day, it really comes down to the developers behind Kubernetes. They've built a tool that tons and tons of people want to use. And that leverages 15 years of work that Google has done on containerization. Work that IBM and Docker and all of our other member companies, RedHat, have put together. And now, I think tiger by the tail is the right analogy. That we just happen to be, luckily, do have the technology and the constellation technology that a lot of folks want to do. The biggest thing we're trying to deal with is, some of the challenges around scaling. There's over 17 hundred authors. Individual developers contributed to Kubernetes in the last 12 months. Trying to figure out how can we get good reviews of all their codes, better documentation. >> There is a secret formula if you look at it. In away, relevance is one of them. >> Dan: Umhmm. >> Being relevant and being an awesome technology. But what I want get your thoughts in is, I looked at Kubernetes right out of the gate and said, hmm, will this be a MapReduced moment for Google. >> Dan: Yeah. >> And interesting enough, they didn't pull the same move. They didn't just let Cloud Air, walk away with or someone. >> Dan: Right, exactly. >> They made sure that if they preserved it. Google kind of let MapReduced >> Dan: Yeah, I think-- >> on the side of the road. >> Dan: No, no, I think this-- >> Cloud Air ran with it. >> Google had something that they replaced it with. I mean the -- >> SPAN is pretty damn good. >> And that's an interesting thing because in a world of strategy, across technology, and this is related to this, is that it used to be, you define a process, and then let's call it the end level process, and then you would go off and you make it obsolete because you had something that was more efficient, more effective. And then you license the old technology. And that way, the industry built capacity around the old technology and you had the new, more efficient technology that drove your business forward. And I think that, I'm not saying that's exactly, I'm not saying that Google did that, that's the tremendous >> Google knew. >> effect it will have. >> John: I have sources that tell me that. I investigated this story three years ago or maybe four, maybe three years ago. Google had conversations going up to the Eric Schmidt level, and Larry Page level, do we keep Kubernetes, do we open source it? And it went all the way to the top. And they almost wanted, they were afraid of MapReduced. Because MapReduced was a lost opportunity. Now they made it up but-- >> Now I would argue that there's a slightly subtler decision they had to make, where they have this internal system board, that is just tons of engineering and analysis and improvement has gone into it. They wrote Kubernetes as essentially next generation version of that. I think they kind of had four paths. Craig McLuckie was one of the key people behind that. Where they could have made it a proprietary service that if you're a customer of Google cloud, you get access to it. That's essentially what Amazon and Elastic Container Services today. Or they could have said, hey, we're going to open source it but we're still keep control of it. Essentially that's the path they went with the Go language. Where lots of people use it, lots of people contribute to it, but it's Google who decides at the end of the day, which direction it goes. Or they could have gone and created a Kubernetes Foundation. And if they'd gone to the Linux Foundation and said, we want to create a Kubernetes Foundation, they absolutely could have and that would have been a home for it. But when you look at all the complementary technologies that have come in, they would never have gone into a Kubernetes Foundation. So instead, they really chose the most open path of saying, no we want to have a Cloud Native Computing Foundation. Have Kubernetes be the anchor tenant for it. But then have a place that companies like Mesophere with Mesos and Docker with Docker Swarm and other partners can come in and agree on something. So today, we're really pleased to announce the container network interface, just got accepted as our 10th project. And that's used by those and also by Cloud Foundry. And then they can disagree on others, about the orchestration- >> So it's a liberating move, really, if you think about it. Because at the time this happened, there was a lot of land grab talk going on. >> Dan: Umhmm. >> Until Amazon was winning big the hockey stick was going up. >> Dan: Right. You saw the numbers, and financial performance. But there was a fear of lock-in. To your point. >> Dan: Right, exactly. >> Then Kubernetes provides a nice layer. And you guys as a group, are looking holistically and saying, choice and multi-cloud. Is that the vision? >> Definitely. But, I mean you can see, strategically why Google decided to do it. Because if you pick an open source platform, and say, hey, this is the best of breed approach. Now, you're actually willing to evaluate the cloud on what the prices are, the supplementary services, et cetera. Where before that, you might have just said, ah, AWS is the safe service, I'm going just go with that. >> But Kubernetes is an invasive technology. And I don't mean that in a bad way. (Dan laughs) >> When you decide to move with Kubernetes, you are foreclosing other options at your disposal. And so, I think what you're saying is that, Google wanted to ensure that it remained a consistent coherent thing. While at the same time, making it obvious to all those around them that also wanted to invest in it, that their investments were going to be safe and sound going forward. >> I think that's fair but on the other hand, I do want to say that very few companies have moved their entire business and all of their IT over to Kubernetes. >> Peter: Oh, I'm not saying that they would. >> We do recommend that they start with a stable service. >> Peter: But Meso and some of those other companies are now investing in Kubernetes as a platform. Or making a bet on Kubernetes, want to make sure that their bets are as good as their company is. >> Sure. But there are other orchestration plateforms still. So Kubernetes has plenty of competition. And our biggest competition of course is Enertia. Of folks not changing into anything. >> I got to ask you a question. So Leonard, our producer is just telling me, Kubernetes is boring per Craig McLuckie. So Craig said earlier in theCUBE today, Kubernetes needs to be boring. He said his biggest problem with Kubernetes is it's too exciting right now. >> Dan: That's great. Now what he means by that is, he's kind of making a play on words but his point is, it should be obstracted away. >> Dan: Yeah. In terms of Kubernetes. But that's a problem you have. It's too exciting. >> Dan: Umhmm. What's your reaction to his comment that Kubernetes needs to be boring. >> He and I did a little Google trends comparison of Kubernetes and TensorFlow, which is another open source project out of Google. TensorFlow is something like three or four acts. And artificial intelligence is just so much more interesting and exciting. And yeah, I certainly would love to see a situation. We have this metaphor for Linux, with the Linux Foundation. That we describe it as plumbing. Where it's so intrinsic to almost every piece of technology in existence. And like plumbing, you'll get very upset when if it stops working. And you'll know it and you'll complain. But there's a huge piece of what we're trying to do which is the infrastructure to make things work. >> Here's an idea. Marketing idea. Just call it AI for containers. >> Dan: That's good. >> It'll be the hottest thing on the planet. >> Dan, great to-- >> Peter: Probably be more be more exciting. >> Dan, great to see you. Congratulations on your success. >> Yeah. So I do want to just make a quick mention December sixth through eighth is CloudNativeCon and KubeCon. It's our biggest annual conference. We're looking to actually triple in size from Seattle to three thousand people or more. We have every expert coming in. Michelle Noorali and Kelsey Hightower are the co-chairs and are going to be speaking there. We would love to see a lot of you guys. >> John: In Austin. >> In Austin. >> We hope you'll be there. >> TheCUBE will be there. >> We'll definitely be there. >> Dan: As well to ah, >> We've been to the inaugural >> Dan: Exactly. >> show for KubeCon and Cloud Native conference. We'll defintely be there. December sixth through the eighth, in December, in Austin. Great time of the year to be in Texas. Congratulations on all your success. And as Kubernetes and nine other projects continue to get traction. Still exciting times. And as they say, we live in interesting times. (Dan laughs) This is theCUBE with more interesting, exciting, not boring stuff coming back from the inaugural event here at Cisco DevNet Create. I'm John Ferrier, Peter Burris. Stay with us.

>> Narrator: Live from San Jose, California It's the CUBE, covering Big Data Silicon Valley 2017. >> Okay welcome back everyone, live in Silicon Valley, this is the CUBE coverage of Big Data Week, Big Data Silicon Valley, our event, in conjunction with Strata Hadoop. This is the CUBE for two days of wall-to-wall coverage. I'm John Furrier with Analyst from Wikibon, George Gilbert our Big Data as well as Peter Buress, covering all of the angles. And our next guest is Wei Wang, Senior Director of Product Market at Hortonworks, a CUBE alumni, and Oliver Chiu, Senior Product Marketing Manager for Big Data and Microsoft Cloud at Azure. Guys, welcome to the CUBE, good to see you again. >> Yes. >> John: On the CUBE, appreciate you coming on. >> Thank you very much. >> So Microsoft and Hortonworks, you guys are no strangers. We have covered you guys many times on the CUBE, on HD insights. You have some stuff happening, here, and I was just talking about you guys this morning on another segment, like, saying hey, you know the distros need a Cloud strategy. So you have something happening tomorrow. Blog post going out. >> Wei: Yep. >> What's the news with Microsoft? >> So essentially I think that we are truly adopting the CloudFirst. And you know that we have been really acquiring a lot of customers in the Cloud. We have that announced in our earnings that more than a quarter of our customers actually already have a Cloud strategy. I want to give out a few statistics that Gardner told us actually last year. The increase for their end users went up 57% just to talk about Hadoop and Microsoft Azure. So what we're here, is to talk about the next generation. We're putting our latest and greatest innovation in which comes in in the package of the release of HDP2.6, that's our last release. I think our last conversation was on 2.5. So 2.6's great latest and newest innovations to put on CloudFirst, hence our partner, here, Microsoft. We're going to put it on Microsoft HD Insight. >> That's super exciting. And, you know, Oliver, one of the things that we've been really fascinated with and covering for multiple years now is the transformation of Microsoft. Even prior to Satya, who's a CUBE alumni by the way, been on the CUBE, when we were at XL event at Stanford. So, CEO of Microsoft, CUBE alumni, good to have that. But, it's interesting, right? I mean, the Open Compute Project. They donated a boatload of IP into the open-source. Heavily now open-source, Brendan Burns works for Microsoft. He's seeing a huge transformation of Microsoft. You've been working with Hortonworks for a while. Now, it's kind of coming together, and one of the things that's interesting is the trend that's teasing out on the CUBE all the time now is integration. He's seeing this flash point where okay, I've got some Hadoop, I've got a bunch of other stuff in the enterprise equation that's kind of coming together. And you know, things like IOT, and AIs all around the corner as well. How are you guys getting this all packaged together? 'Cause this kind of highlights some of the things that are now integrated in with the tools you have. Give us an update. >> Yeah, absolutely. So for sure, just to kind of respond to the trend, Microsoft kind of made that transformation of being CloudFirst, you know, many years ago. And, it's been great to partner with someone like Hortonworks actually for the last four years of bringing HD Insight as a first party Microsoft Cloud service. And because of that, as we're building other Cloud services around in Azure, we have over 60 services. Think about that. That's 60 PAZ and IAZ services in Microsoft, part of the Azure ecosystem. All of this is starting to get completely integrated with all of our other services. So HD Insight, as an example, is integrated with all of our relational investments, our BI investments, our machine learning investments, our data science investments. And so, it's really just becoming part of the fabric of the Azure Cloud. And so that's a testament to the great partnership that we're having with Hortonworks. >> So the inquiry comment from Gardner, and we're seeing similar things on the Wikibon site on our research team, is that now the legitimacy of say, of seeing how Hadoop fits into the bigger picture, not just Hadoop being the pure-play Big Data platform which many people were doing. But now they're seeing a bigger picture where I can have Hadoop, and I can have some other stuff all integrating. Is that all kind of where this is going from you guys' perspective? >> So yeah, it's again, some statistics we have done tech-validate service that our customers are telling us that 43% of the responders are actually using that integrated approach, the hybrid. They're using the Cloud. They're using our stuff on-premise to actually provide integrated end-to-end processing workload. They are now, I think, people are less think about, I would think, a couple years ago, people probably think a little bit about what kind of data they want to put in the Cloud. What kind of workload they want to actually execute in the Cloud, versus their own premise. I think, what we see is that line starting to blur a little bit. And given the partnership we have with Microsoft, the kind of, the enterprise-ready functionalities, and we talk about that for a long time last time I was here. Talk about security, talk about governance, talk about just layer of, integrated layer to manage the entire thing. Either on-premise, or in the Cloud. I think those are some of the functionalities or some of the innovations that make people a lot more at ease with the idea of putting the entire mission-critical applications in the Cloud, and I want to mention that, especially with our blog going out tomorrow that we will actually announce the Spark 2.1. In which, in Microsoft Azure HD Insight, we're actually going to guarantee 99.9% of SLA. Right, so it's, for that, it's for enterprise customers. In which many of us have together that is truly an insurance outfield, that people are not just only feel at ease about their data, that where they're going to locate, either in the Cloud or within their data center, but also the kind of speed and response and reliability. >> Oliver, I want to queue off something you said which was interesting, that you have 60 services, and that they're increasingly integrated with each other. The idea that Hadoop itself is made up of many projects or services and I think in some amount of time, we won't look at it as a discrete project or product, but something that's integrated with together makes a pipeline, a mix-and-match. I'm curious if you can share with us a vision of how you see Hadoop fitting in with a richer set of Microsoft services, where it might be SQL server, it might be streaming analytics, what that looks like and so the issue of sort of a mix-and-match toolkit fades into a more seamless set of services. >> Yeah, absolutely. And you're right, Hadoop and Wei will definitely reiterate this, is that Hadoop is a platform right, and certainly there is multiple different workloads and projects on that platform that do a lot of different things. You have Spark that can do machine learning and streaming, and SQL-like queries, and you have Hadoop itself that can do badge, interactive, streaming as well. So, you see kind of a lot of workloads being built on open-source Hadoop. And as you bring it to the Cloud, it's really for customers that what we found, and kind of this new Microsoft that is often talked about, is it's all about choice and flexibility for our customers. And so, some customers want to be 100% open-source Apache Hadoop, and if they want that, HD Insight is the right offering, and what we can do is we can surround it with other capabilities that are outside of maybe core Hadoop-type capabilities. Like if you want to media services, all the way down to, you know, other technologies nothing related to, specifically to data and analytics. And so they can combine that with the Hadoop offering, and blend it into a combined offering. And there are some customers that will blend open-source Hadoop with some of our Azure data services as well, because it offers something unique or different. But it's really a choice for our customers. Whatever they're open to, whatever their kind of their strategy for their organization. >> Is there, just to kind of then compare it with other philosophies, do you see that notion that Hadoop now becomes a set of services that might or might not be mixed and matched with native services. Is that different from how Amazon or Google, you know, you perceive them to be integrating Hadoop into their sort of pipelines and services? >> Yeah, it's different because I see Amazon and Google, like, for instance, Google kind of is starting to change their philosophy a little bit with introduction of dataproc. But before, you can see them as an organization that was really focused on bringing some of the internal learnings of Google into the marketplace with their own, you can say proprietary-type services with some of the offerings that they have. But now, they're kind of realizing the value that Hadoop, that Apache Hadoop ecosystem brings. And so, with that comes the introduction of their own manage service. And for AWS, their roots is IAZ, so to speak, is kind of the roots of their Cloud, and they're starting to bring kind of other systems, very similar to, I would say Microsoft Strategy. For us, we are all about making things enterprise-ready. So that's what the unique differentiator and kind of what you alluded to. And so for Microsoft, all of our data services are backed by 99.9% service-level agreement including our relationship with Hortonworks. So that's kind of one, >> Just say that again, one more time. >> 99.9% up-time, and if, >> SLA. >> Oliver: SLA and so that's a guarantee to our customers. So if anything we're, >> John: One more time. >> It's a guarantee to our customers. >> No, this is important. SLA, I mean Google Next didn't talk much about last week their Cloud event. They talked about speed thieves, >> Exactly >> Not a lot of SLAs. This is mandate for the enterprise. They care more about SLA so, not that they don't care about price, but they'd much rather have solid, bulletproof SLAs than the best price. 'Cause the total cost of ownership. >> Right. And that's really the heritage of where Microsoft comes from, is we have been serving our on-premises customers for so long, we understand what they want and need and require for a mission-critical enterprise-ready deployment. And so, our relationship with Hortonworks absolutely 99.9% service-level agreement that we will guarantee to our customers and across all of the Hadoop workloads, whether it would be Hive, whether it would be Spark, whether it'd be Kafka, any of the workloads that we have on HD Insight, is enterprise-ready by virtue, mission-critical, built-in, all that stuff that you would expect. >> Yeah, you guys certainly have a great track record with enterprise. No debate about that, 100%. Um, back to you guys, I want to take a step back and look at some things we've been observing kicking off this week at the Strata Hadoop. This is our eighth year covering, Hadoop world now has evolved into a whole huge thing with Big Data SV and Big Data NYC that we run as well. The bets that were made. And so, I've been intrigued by HD Insights from day one. >> Yep. >> Especially the relationship with Microsoft. Got our attention right away, because of where we saw the dots connecting, which is kind of where we are now. That's a good bet. We're looking at what bets were made and who's making which bets when, and how they're panning out, so I want to just connect the dots. Bets that you guys have made, and the bets that you guys have made that are now paying off, and certainly we've done before camera revolution analytics. Obviously, now, looking real good middle of the fairway as they say. Bets you guys have made that hey, that was a good call. >> Right, and we think that first and foremost, we are sworn to work to support machine learning, we don't call it AI, but we are probably the one that first to always put the Spark, right, in Hadoop. I know that Spark has gained a lot of traction, but I remember that in the early days, we are the ones that as a distro that, going out there not only just verbally talk about support of Spark, but truly put it in our distribution as one of the component. We actually now in the last version, we are actually allows also flexibility. You know Spark, how often they change. Every six weeks they have a new version. And that's kind of in the sense of running into paradox of what actually enterprise-ready is. Within six weeks, they can't even roll out an entire process, right? If they have a workload, they probably can't even get everyone to adopt that yet, within six weeks. So what we did, actually, in the last version, in which we will continue to do, is to essentially support multiple versions of Spark. Right, we essentially to talk about that. And the other bet we have made is about Hive. We truly made that as kind of an initiative behind project Stinger initiative, and also have ties now with LAP. We made the effort to join in with all the other open-source developers to go behind this project that make sure that SQL is becoming truly available for our customers, right. Not only just affordable, but also have the most comprehensive coverage for SQL, and C20-11. But also now having that almost sub-second interactive query. So I think that's the kind of bet we made. >> Yeah, I guess the compatibility of SQL, then you got the performance advantage going on, and this database is where it's in memory or it's SSD, That seems to be the action. >> Wei: Yeah. >> Oliver, you guys made some good bets. So, let's go down the list. >> So let's go down memory lane. I always kind of want to go back to our partnership with Hortonworks. We partnered with Hortonworks really early on, in the early days of Hortonworks' existence. And the reason we made that bet was because of Hortonworks' strategy of being completely open. Right, and so that was a key decision criteria for Microsoft. That we wanted to partner with someone whose entire philosophy was open-source, and committing everything back to the Apache ecosystem. And so that was a very strategic bet that we made. >> John: It was bold at the time, too. >> It was very bold, at the time, yeah. Because Hortonworks at that time was a much smaller company than they are today. But we kind of understood of where the ecosystem was going, and we wanted to partner with people who were committing code back into the ecosystem. So that, I would argue, is definitely one really big bet that was a very successful one and continues to play out even today. Other bets that we've made and like we've talked about prior is our acquisition of Revolution Analytics a couple years ago and that's, >> R just keeps on rolling, it keeps on rolling, rolling, rolling. Awesome. >> Absolutely. Yeah. >> Alright, final words. Why don't we get updated on the data science experiences you guys have. Is there any update there? What's going on, what seems to be, the data science tools are accelerating fast. And, in fact, some are saying that looks like the software tools years and years ago. A lot more work to do. So what's happening with the data science experience? >> Yeah absolutely and just tying back to that original comment around R, Revolution Analytics. That has become Microsoft, our server. And we're offering that, available on-premises and in the Cloud. So on-premises, it's completely integrated with SQL server. So all SQL server customers will now be able to do in-database analytics with R built-in-to-the-core database. And that we see as a major win for us, and a differentiator in the marketplace. But in the Cloud, in conjunction with our partnership with Hortonworks, we're making Microsoft R server, available as part of our integration with Azure HD Insights. So we're kind of just tying back all that integration that we talked about. And so that's built in, and so any customer can take R, and paralyze that across any number of Hadoop and Sparknotes in a managed service within minutes. Clusters will spin up, and they can just run all their data science models and train them across any number of Hadoop and Sparknotes. And so that is, >> John: That takes the heavy lifting away on the cluster management side, so they can focus on their jobs. >> Oliver: Absolutely. >> Awesome. Well guys, thanks for coming on. We really appreciate Wei Wang with Hortonworks, and we have Oliver Chiu from Microsoft. Great to get the update, and tomorrow 10:30, the CloudFirst news hits. CloudFirst, Hortonworks with Azure, great news, congratulations, good Cloud play for Hortonworks. To CUBE, I'm John Furrier with George Gilbert. More coverage live in Silicon Valley after this short break.

>> Computer Museum, in the heart of Silicon Valley, extracting the signal from the noise. It's theCUBE. Covering OpenStack Silicon Valley 2015. Brought to you by Mirantis. Now, your hosts John Furrier and Jeff Frick. (upbeat music) >> Okay welcome back everyone. We are here live, broadcasting. This is SiliconANGLE Media, theCUBE, our flagship program. We go out to the events and extract the signal from the noise. I'm John Furrier, my co-host Jeff Frick this week. Two days of wall-to-wall coverage live in Silicon Valley for OpenStack Silicon Valley or #OpenStackSV or the hashtag for this event today, #OSSV15. Join the conversation. Join our crowd chat, crowdchat.net/OSSV15. Our next guest is Craig McLuckie, who's with Google. He's on the Google Cloud team, CUBE Alum. Welcome back to theCUBE. Got a keynote there, welcome back. >> Thank you so much, great to be with you again. >> So Silicon Valley house leads center of the innovation engine house a lot of investment capital here, a lot of big players, you guys, Facebook, VMware, Intel, you name it. It's the giants of the technology industry. And the bubble conversation's happening. China's going down in terms of economics, and seeing the stock market crash there. But yet, underlying infrastructure change is happening. Cloud certainly is floating, that wealth-creation engine, you guy are a big part of it here in Silicon Valley. Just talk about the state of the Cloud. OpenStack has momentum, you have some stability in the core compute side with OpenStack, virtualization is not going away. New things like Kubernetes, Containers, fast on the scene, rising very fast. What's your take on this innovation engine in the Cloud? >> So I think there's a couple of things that are really exciting and interesting that are happening right now, as we speak. The first is a transition to open. It's a way of rethinking about how you evaluate, acquire, and integrate your software. And I think that OpenStack has established a legitimacy as a technology that's really bringing the value proposition of traditional infrastructure service to everyone everywhere. And we're really starting to see a convergence to that community, a set of technologies that are consistent, of high simatic consistency, is really becoming a thing, which is phenomenal. At the same time we're also seeing another disruption happening. And it was really a disruption that was triggered by the emergence of Docker as a technology to support a new way of thinking about packaging and deployment. And it's really part of a bigger story around a move towards Cloud -Ntive computing. This is a computing set of patterns that was really inspired by the internet giants by the Google's, the Facebook's, the Twitter's. But it's really been cracked open and been accessible by folks like Docker who have opened up those container technologies and now we're seeing a lot of the players start to really focus on this and look at bringing the value proposition of that new style of computing to enterprises everywhere. >> You know you start to see maturity in a market specially when platforms are involved, platform wars, whatever the bloggers want to put the headline out there, when you see abstraction layers develop. And one of the things that you talked about in your keynote I'd like you can elaborate on is ending the distinction between what's under the hood. Containers you mentioned bring out this notion that, "I'm a developer I want interoperability." >> Right. >> "I want cross platform API's." This is the economy so I want you to explain that. What is this disruption with containers and Kubernetes? Do, for this abstraction, do we care about the features any more? And that's one of the signals of maturity. Is that you're not talking speeds and feeds and infrastructure to service, platform as a service. When those conversations go away you know things are moving. >> Right. >> Or is that true, what's your take on all that? >> I think that's a very good observation. I think that one of the things we as a community have looked for for a while is a separation between the world of tools and infrastructure that people interact with on a day to day basis to build applications and in the systems that actually take those built applications and run them for you. And a big part of our focus has been to make the set of subsystems that are actually responsible for the operations of applications, transparent to the end developer. And we're looking to formalize that interface that exists between how you create an application, how you package up it's dependencies and how you offer up the infrastructure and then how you run it. One of the most exciting and energizing things for me is to see the emergence of a standard set of abstraction that interface between these two worlds so it creates massive opportunities for innovation. By standardizing that interface you have incredible innovation in the tooling area with technologies like Docker or continuous integration of delivery frame works. You know new development environments that are producing an artifact that can be universally consumed everywhere else. And then on the infrastructure side you have a lot of innovation around running that artifact for the developer, the end enterprise efficiently and intelligently whether it's being deployed into a virtual machine on OpenStack with being deployed into a main-source cluster running on the metal or whether it's been deployed into a next generation Kubernetes cluster running in one of those environments or somewhere else. We're looking to create this common abstraction and it's going to drive a lot of innovation at every level of the stack. >> You know at Wikibon research one of the things that they're putting out, some cutting edge research around the innovation around some of the technologies under the hood. Conversion infrastructure, cloud technologies, flash, storage, software defined networking all that stuff under the hood is evolving as fast as well. So you have underlying core technology and tooling exploding. >> Right. >> So some really good stuff coming out Wikibon.com. And with that and your comment I want to ask, kind of a pointed question which is: Does hybrid cloud really exist? Is it a concept or is it a category? Do people buy hybrid-cloud? Do they buy into it? It seems to be that's the conversation people are talking about now. But I just don't see hybrid-cloud existing other than being part of private and public. >> Right. >> And talk about that. >> It's a great question. I love that question. It exists but not the way that people think of it existing. Right so you can think about it this way when you are building an application on your laptop and deploying it into a cloud it's kind of hybrid-cloud right? But it's not the way that people think about hybrid-cloud. When you want to run a continuous integration server for your company and have it hosted in the cloud and have it create artifacts that are deployed into you on-prem production clusters. That's hybrid-cloud but it's not the way people have come to think about it. And so what I think about it is really about the ecosystem. About establishing a common set of tools and capabilities so that first and foremost people can choose the destination for an application based solely on the technical merits of the infrastructure that they're ruing on. Google offers some very high quality, robust, fast, affordable cloud infrastructure. But we recognize and embrace the fact that for some customers you have very legitimate regional requirements. For some of the applications you might really want to run them on premises. And so the first step toward achieving legitimacy for hybrid-cloud is establishing a common set of patterns and tools and capabilities that exist in both places. The next step is going to be around creating a common services abstraction that let's you start to access things from other environments. And then over time you might actually see people deploy these sort of cloud bursting scenarios et cetera. But the path to get there is really through infrastructure. You know like a common set of abstractions, a common set of tools, a common set of pattern, and making those available to people everywhere. And then over time we will start building these fused together, legitimately hybrid solutions. >> So hybrid-cloud then is a paradigm, it's a concept that highlights the common tooling interoperability so developers can actually work in these environments without having to do anything. That's where Docker comes in, that's where Kubernetes come in? >> Exactly. And it's really, hybrid needs to be first and foremost about being able to use a common set of technologies to build an application for A or B. >> So let's take it forward. So let's put the brainstorming hat on. Let's talk about the future and let's kind of play with some scenario's. Internet of things opens up a huge can of worms and challenges, engineering challenges around: How do I manage the data? How do I drive workloads to these devices? Whether their wearables or cars or stacks or devices? Anything that's on the edge of the network is now considered a device. PC, mobile, internet of things. So for a developer to work in that kind of environment they need these toolings. Is that how you see it? >> Absolutely, I think that's a great way to think about it. You know it's an interesting thing you raise. Because if you think about it Cloud-Native has really been the domain of internet companies, right? It's really been something that Google's done because it's the only way to practically achieve a certain level of scale. We've seen co-evolution of this, of these patterns inside Twitter, eBay, Facebook, Netflix. Everyone's been doing it on their own terms. Now the reality is when IoT happens every enterprise has to kind of become a internet company right? And what we've seen consistently across, you know all of the internet companies that exist today is there's one pattern that really works well to actually deploy computational infrastructure, at scale efficiently. And that's this pattern around container package, dynamically schedule, microservices oriented computing. And so our mission is really to bring these technologies in a democratized way to enterprises so that they can actually tackle problems that were previously only really solved by the internet giants. Without having to make Google level investments or Facebook level investments in technology. >> Yeah. When we hear that Internet companies, just clarify like a hyperscaler like with Yahoo and Google did. Building large scale systems in a seamless way that's kind of abstract to the user. >> Right. >> Just pure performance all, everything is running and it's kind of a brilliant concept. That brings up the point of Google envy. I mean you hear this all the time in the enterprise. "I want to be more like Google." "I want to be more like Facebook." And what they really are saying is: "I want to have Ops." Right so. >> Right. >> DevOps, Cloud-Native do you hear hat often? And when you hear that: "I want to be more like Google." What does that really mean from your stand point? How do you guys internalize that? >> Right. >> How do you talk back to customers? >> So I think you know when I say I want to be more like Google I think there's a lot of different sort of angles that you might have there. I've heard people coin this phrase GIFEE to describe what we're trying to do: Google Infrastructure for Everyone else. But I think the heart of it is really this: If you're a Google engineer, it's like you have a superpower, right. You have access to this amazing almost unlimited mass of infrastructure that's just at your disposal immediately. At very little cost or overhead. And you don't have to worry about the mechanics of actually where the thing I built is run, right. Operations is just a function of the platform. The developer gets to focus on their application and their application operations and what they get for free is this cluster environment where cluster operations is handled for you. The process of actually mapping an atom of code into a distributors systems environment. The ability to use some very powerful services that make it trivial to build distributable systems. The fact that I'm not paged all the time because what I deploy is understandable by some very smart subsystems, they can watch it, they know what it's supposed to be doing. They can tell when it's not doing that and they know how to fix it, right. And so traditionally when you go out of operating parameters in a traditional system you get paged. And for me a lot of what this operate like Google really means is one is I want to be able to Access Compute at an unprecedented level easily and two is I don't want to get paged by my applications that are doing that. >> Yes so let's bring that up, the API economy. Let's bring this to the next level. Today applications are either Legacy or their Cloud-Native so and I ask everyone the question, even on our own Wikibon team we have a debate. And I ask Dave Alante: "Dave name the Cloud-Native Apps that are out there?" I don't think there are any Cloud-Native apps out there. I mean who has a Cloud-Native App? Now that's a trick question because he goes: "Amazons an App, Google has Cloud-Native." Well they're already hyperscaled. >> Right. >> So the question is what, where are the Cloud-Native apps? Where are the examples? Now Facebook's a Cloud-Native App because they built it from the ground up to be Cloud-Native. >> Sure. >> Google same way. So as an enterprise, what is the Cloud-Native App to the enterprise and how do they get there? And what Legacy do they have to throw away because its synchronous and API interactions is fundamental. >> Right. >> How do you ease that out? >> This is actually a fascinating topic and I think one of the most dangerous things people assume is that to accomplish Cloud-Native you have to go fully along the API-Fication path, right. Now the reality is that of you look at they way that people access data today the fast majority of business data's stored in relational database's. People have great tools to access data in relational databases. They want to be able to move that forward. And to me if you force API-Fication, if you force a protocol specific approach to actual integration, if you force people to use a specific authentication scheme you're going to alienate a very broad array of your customers and you're going to create this cognitive hurdle that's very hard for people to get over. So when I think about Cloud-Native, I think about it as providing a different paradigm for deployment management, activation et cetera. But it has to make allowances for integration with your existing systems. And so I think at the forefront of this is the notion of a service or a microservice. And a microservice has to be a minimal atom of software consumption, the easiest way to find and consume something and you can't force an opinion around how people project that, right. So if you build something that runs in a cluster you should be able to access an Oracle database as if it were a microservice running inside your cluster. You should be able to access a sales force SAS endpoint as if it were a microservice running inside your cluster. And so as I think about my mission and Google's mission around the move towards Cloud-Native computing, you can't create this experiential cliffs, you can't create these artificial boundaries to your system. You have to make natural allowances where, look there's some stuff that just works better in a vertically scalable VM. If you want to run a big database with a Dune Kernel and a few other things, by all means put it in a VM. And we are absolutely committed to the idea of creating a natural set of experiences when you want to go from that to some portion of the application that's doing stateless, front and serving. Or a portion of the application that's running in a cloud-friendly, distributed scaled out database. You shouldn't have to take the pull and be stuck in this world. You should be able to mix these. >> So you're saying it's dangerous to force API-Fication, if that's a term I can't even spell it, It's too may I's at the end there, I like that hyphen in there. But if you force API-Fication or movement, you can foreclose future performance and functionality by alienating existing apps. >> By alienating the existing system. It is a very dangerous, there's a lot of. It's very attractive to drive API-Fication but it has to be, you have to create this pressure grading that attracts people up it by adding value at every stage of the game. And you can't build your management systems around a predicated, sort of, opinionated API framework. We saw this with, in the world of SOA, I mean I don't know if you remember the SOAP and SOA stuff. >> Yeah, yeah. >> You know way back when. >> That was just another way of describing API-Fication and we've saw where it went. The problem was that. >> It wasn't ready, the market wasn't ready for web services at that time. >> And it was, but it was beyond that, it was like, no one's willing to make a massive infrastructure investment to get you to ground zero, where you can actually start building. >> So let's look at that web services back in 2000, 2001 when you saw SOAP, XML, SAM all those things emerging. At that time who did take advantage of that? It was the hyperscalers. It was internet companies cause they needed it, right. So the mainstream market now is adopting that kind of concept around microservices. Explain that. >> But it wasn't, the interesting thing is when you look at what the adoption was around microservices, it wasn't around interoperable SOAP, it was around discrete, highly optimized RPC protocols. It was around relatively closed systems at that time. And it worked well, right? The challenge. >> It was controlled. >> It was controlled and it worked well inside a closed ecosystem. Now what, the thing that really held people back is that to get there you had to do a big ESP deployment. You had to then go and SOA-fy a bunch of your components and it required a huge investment in terms of sort of infrastructure and capabilities to get, before you started realizing value. And it was inaccessible to most people and it alienated technologies that didn't fit well into that model. Right like how do you take your database and put it into that model? It was purely optimized around a certain portion of it. And so now we're in a world where we make it available to everyone. We reduce barriers to entry and you get immediate value without having to make huge investments. So let's take microservices and let's unpack that for the audience out there. You're seeing DockerCon, ContainerCon, KubeCon, MasosCon. All these conferences are around developers. And this is all about scale right? >> Right. >> Operating a scale, abstraction layers. I think it's, we need to be careful not to pigeonhole this as about operating at scale. It is the only practical way to operate at internet scale but the value proposition is just as applicable if you're running something in five virtual machines, at a more humble scale. >> So let's talk about development versus operation team. >> Right. >> Where does the Kubernetes, where does the microservices model fit in? And how do companies avoid the trap of alienating existing apps? How do they get the system up and running? What is the roadmap? And differentiate from a Dev standpoint and an Ops standpoint. >> I think one of the most important things you're going to start seeing is a specialization of the operations function. Today it's all kind of glum together and if you ask a developer to actually run an application they have to be cognizant of which virtual machine it's in. You force them into the ugly world of infrastructure Ops. And sort of common services Ops. And what we're going to start seeing, and what I hope to help companies achieve, is a specialization of the operations function. So Infrastructure Ops should be relegated to a set of people that actually understand the physical infrastructure. They will create an optimal physical environment surround your application. There'll be a small number of specialized people that know how to do that and they will rack and stack and wire and configure and do what ever needs to be done to tune the infrastructure. Above that you're going to see this Cluster Operations. So a common Services Operation team that provide a basic operational platform and common services to everyone. So these are a highly specialized set of people that provide you the tools you need to be able to autonomously run a distribute system. They are unlikely to be involved in the day to day operations because most of these systems will be autonomous but they're there to answer the call if something happens in, in that system. So it becomes a very specialized function. And Google does this with our SRE folks that actually manage our, like the Boar clusters that run all our infrastructure. Small number of highly specialized people providing a very valuable service to a lot of folks. And then at the top level you're going to have Application Operations. And that really just becomes the developers function. And it should really be about understanding and managing your code and you should never have to think about: Where it's running? How it's running? You never, should never have to SSH into an instance to try and debug it. All that should be presented to you through your tools. So the developer's experience becomes one of of using logical infrastructure. And so I think what we're going to start seeing is companies making investments in these clustering technologies. Offering up these simple, clustered service environments for their departments. And then having portfolio's of container package applications that can be easily taken, adjusted and run in these environments. And we'll naturally see the specialization of operations emerge. >> So we're running out of time. Jeff didn't get one question in but maybe next time. >> He has a role in that. >> Brendan Burns, Brendan Burns I think on your team. >> Yeah. >> Brendan, so he brought up something. He brought up the hybrid-cloud is kind of the way, meaning the way you described it, not as a category. But he also brought up the different aspects of Google Cloud in our last crowd chat last month. How do customers mix and mach with the cloud? I mean you guys offer Linux, you guys offer Windows. I mean if I want to work with Google Cloud what are the touch points? How do people ingratiate in? How do they engage with Google? What are some of the use cases? Can you share just put the plug in for Google Cloud what you guys have up and running that's mature, stable, >> Right >> Shipping. And how do customers get into the Google Cloud? >> So we've really seen Google Cloud, it needs to be in all of the above sort of capabilities. The operating characteristics. The thing that make Google Cloud unique is the quality of the basic infrastructure. We offer by far the most price performing basic infrastructure out there. It's an innovative clouds, you know it's driving and active in a lot of the sort of disruptions we're seeing around the container space. It's an open cloud. It's a cloud that's invested in making sure that we engage and connect with the OpenSource community. So if you want to work with Google Cloud there's a lot of different ways to do it. One is you can go and just buy beautiful, clean, pristine, powerful, affordable infrastructure in large chucks through Google compute engine. And we're seeing a tremendous amount of adoption. You don't have to make massive capex down payments to get our best price. We really focus on doing that. You can also come in if you just want to write a bit of code, have it run, we have a wonderful Pass product called Google App engine that's becoming very naturally integrated into the container ecosystem and is a natural sort of path. It's a great entry point for people that just want to operate on a higher level and want to take some code and then have it easily deployed and run on your behalf. And then we're also, another entry pointy that isn't obvious to people is, you can help us build the Google Cloud. What we're building with our next generation set of offerings with technologies like Google Container Engine, is an opensource cloud. It's been built in public. Come join our community, work with us. Try it out. Give us feedback and be part of actually building the next generation of clouds. >> Okay so the question I have for you is, let's just say I'm an Amazon customer and I want to go to Google Cloud, do you have like an elastic Beanstalk application containers an App Engine, how do I get I there? I mean there are some things that Amazon has you might have some things. How do you talk to that, Beanstalk particulars. >> That's a great question. So Beanstalk you know provides the ability to you know deploy and run applications. The closest analogy is App Engine. So Beanstalk traditionally was a Java base platform that you could provide your Java code and it would run it for you. App Engine gives you that equivalent capability. And with the new generation of App Engine we actually provide the ability to deploy into directly into VM. So that it feels a lot, it feels a lot like the Beanstalk experience. But it comes with a lot of other high value services. And so that's a natural starting point. And App Engine it self is being rebased on a lot of the Kubernetes concepts. So that you have this immediate, easy, accessible experience for code but when you reach an edge and you want to actually integrate it naturally with a vertically scaled database that runs in a VM, we have compute engine waiting for you and all very natural, it will feel natural to actually just integrate those two things together and snap together these more holistic solutions. >> You guys have a, final question. I now you guys have a lot of track record with developers certainly Google's history and OpenSource, everything is great. But other competitors, more commercial IBM and Amazon, they're providing marketplaces for distribution, where people can make some cash and some cabbage. >> Right. >> What's the plans Google? Is there anything there? How do I make money if I'm a developer with Google? Or is there plans there, what's the state of that? >> It's a great question and obviously we have aspirations in that space. I can't go into all the details right now. But you know the we are obviously investing in that area. And one of the things that we're really like though is looking at containers as a standard distribution framework, let's you plug into everyone's market places. So one of the things that I see around marketplaces historically is that they offer immediate value in connecting a producer and consumer of software but they're not offering steady state value. So once those two have been connected the marketplace isn't adding significant ongoing value. So when you think about what we want to do, we want to make sure that one is, we become a market maker, we let lost of different market places emerge and that we support those. But then in our own efforts we actually add legitimate value to both the producer and the consumer of the software. And the we're not just taking a cut off the top. So but that's, it will become much more clearer in the face of time. >> Craig, thanks for spending some time and congrats on a great keynote. Good to see you again. Thanks for jumping in and sharing the data here on theCUBE, really appreciate it. We are live here in Silicon Valley. It's theCUBE at OpenStackSV, join the conversation #OSSV15. We'll be right back after this short break. (upbeat music)