>> Narrator: Live, from Las Vegas, it's theCUBE, covering InterConnect 2017, brought to you by IBM. >> Okay, welcome back, everyone. We are live in Las Vegas, at the Mandalay Bay, for IBM's InterConnect 2017. It's the cloud and big-data Watson show that's all kind of coming together. This is theCUBE's three-day coverage, wall-to-wall day two, coming to an end here. I'm John Furrier, with my co-host Dave Vellante. Our next guest is Dave Lindquist who's an IBM Fellow, vice-president of Cloud DevOps and Analytics, at IBM. Great to have you on theCUBE, thanks for joining us. >> Thank you, John, thank you, Dave. >> So, love to have the IBM Fellows on, because we can then, like, get down and dirty, right? Get down and talk about the tech. I don't see if Ginnie's on stage today, I love the bumper sticker she has, 'cause she's, she nails it; enterprise strong, data first, cognitive to the core. So, enterprise strong means, there's a cloud-readiness equation going on right now, and we just came back from Google Next, and, hey, we've got great technology, buy us. Well, SLAs matter. You know, being enterprise ready isn't always about the best tech. >> No, no. >> It's about everything; it's the data, it's the machine-learning, it's the software, and also, those table stakes going on in the enterprise. Unpack that for us. >> Sure. Well, I think a lot of what you just went through, is at least part of the driving force between bringing ops into the dev space, this DevOps thing, and we'll expand on that in a little bit. But one of the big pushes going on is really around site reliability engineering, and how do you appropriately bring the skills together with the development teams to really set systems up in elastic scale, recovery-oriented compute models, so that you can scale that with the demand, you can recover from situations, you can recover from failures, you have a lot of redundancy built in the system. It takes a lot of time for teams to mature, to understand that, that aspect of delivering cloud services and delivering applications into, into a continuous available environment. >> What's IBM's formula for that right now, is you guys ramp-up and scale-up the cloud, IBM Cloud, you have the soft layer, and that's now Bluemix. So you have, on the lower end of the stack, you got to get that hardened infrastructure, if it's a service, and the platforms and service stuff. Then you start to bleed into the Bluemix. It's all one Bluemix now, but, you've got app developers, they want infrastructure as code, they want data as code, but then you got to have an uncoupling of set of services that look like one set of services. How hard is that, and what are you guys doing specifically to talk to customers about the value you're bringing on both sides of that camp? You know, the hard workload focused hybrid, to the creative sizzle of an app. >> Yeah, well, lot in that question, there's a lot of parts, lot of parts there. One of the things that's clearly going on, is, taking that next step in loose coupling systems, creating more independent services that can scale, elastically, independently of each other, the recovery-oriented models, and then presenting those services, up at the layers you mentioned; at the infrastructural areas, compute-storage networking, into the paths and container layers, so that the application developers can very rapidly get the environment they need, compose the services that they need, like the runtimes, data, messaging, et cetera, as a loosely-coupled system, and then build their applications to be deployed into that environment. >> How much innovation is going on? You're starting to see now, a new trend where there's more hardware engineering going into some chips, and hardware configurations, that's essentially software-driven, to offload, maybe machine-learning, some other, cooler things, that can assist some of the hard stuff that frees up more creativity on the software side. Say, machine-learning is a great example, you're starting to see Intel and others start thinking, okay, let's put some stuff on a chip. You have 5G wireless, you've got autonomous vehicles coming, a whole new hardware paradigm is kind of emerging with the cloud; how do you see that playing out, from an innovation standpoint? How does that strategy play out from a cloud, and IoT? >> To me, a lot of the things that are so exciting that's going on in the cloud, probably the big driver in the cloud is this whole acceleration of innovation. How quickly can you get from, instantiate an idea, in-field, iterate in-field with your users, towards a business outcome, and as you hit those outcomes, start scaling and expanding that out. And a lot of that innovation is building on some of the things that you mentioned: big data, cognitive, IoT, social, how do you start bringing these things together? And so, as you bring this together, real-time, you clearly need just exponential growth occurring, in compute capacity, which is probably creating, not probably, it's creating all kinds of opportunities for breakthroughs in algorithms, and breakthroughs in the hardware to support that. >> The other thing that we're seeing, I want to get your thoughts and commentary on is, how analytics is so compatible with the cloud, because, you're seeing that sweet spot developing nicely, and also with cloud-native trend is booming. You're seeing cloud-native compute foundations got big traction, and then the analytics is, people have no problem putting that in the public could, but yet they want the hybrid over here for some other stuff. So the workloads are starting to settle into their swim lanes. Your thoughts on the DevOps equation, as analytics moves to the cloud, not exclusively, but you know, for the majority of cases, and this cloud-native trend that's coming down the pike. >> Yeah, so, break that down in a couple pieces, the cloud-native trend, as well as the analytics trend. The cloud-native trend, what you see is a lot of development with micro services, and part of what makes that so exciting, is the culture of the teams and how they come together. You're basically seeing small teams, small, integrated teams, often called two-pizza teams, or squads, where you pull together designers with developers, with tests, with data science, with business, insights business strategy, into a team that then works together through the whole life-cycle, iterating incrementally and delivering in-field, to, as they move towards that business outcome that they're trying to achieve. So, what cloud-native is doing, is allowing, where that micro service model is really allowing many of these teams to work with relative autonomy, but accountability for their service, as it comes together to bring the full system together. What we're learning is that, one, you get a lot of speed like that, but then you start to, you need a level of analytics to help understand how that's coming together through that whole life-cycle, and what I mean by that is, you know, how is the testing coming along? So that everybody needs to start adopting more continuous testing, from unit tests, right, performance testing, availability, right into security testing. So you start running basic, simple analytics, where you start gathering on how the teams are doing in the continuous testing, and you can start setting soft and hard gates. An example of a soft gate might be, code coverage is dropping, so send an alert to the team to say, you've got to step up the code coverage. A hard gate might be, a security scan failed, so stop the deploy. And so, that's a basic set of analytics, but, the fun areas, to me, the exciting areas, we're starting to apply much more sophisticated models, are in understanding code health, and how the teams are actually working together. So you start developing models-- >> It's almost like team chemistry and coding working together. It's like, hey, you guys are good. You know, you're in the zone, you know? You're in the coding zone. Yeah, but this is a good point, I want to highlight just, let's just stop on that one point, I want to just drill down. I think that you nailed something that's, we've been kind of teasing out, and you put into words, the cloud-native trend around micro services, you mentioned teams working together, maybe some shared analytics, and kind of, code health, team, you know, scoreboard, or whatever. This is way beyond agile. I mean, agile has been a term that's been talked about inside companies, hey, let's be agile. You're talking about a fundamental industry reconfiguration of the players, so this is like a whole 'nother ballgame. >> To me, it builds on agile, what's going on, it does build on-- >> It goes beyond, it's-- >> But it goes way beyond, and even, you know, the early thinking, in DevOps, I think we're really pushing the envelope when we still call it DevOps, because we're thinking of the broad life-cycle of, you know, design practices. How do you begin to understand your users and what you're trying to accomplish with your users? Then you get into, you know, continuous integration, delivery, and testing, but then where it gets real interesting, is you start instrumenting everything, including, you know, getting direct LAN to site insight into how your users are using what you're deploying, and that causes the ability to pivot very rapidly, daily, weekly, into, you know, guiding where you're going to take your next iterations. To me, that's what's really taking this way past what you typically saw in an agile-- >> So what's happening to this traditional IT function? How is it adapting? You know, is it bi-modal, is there subtraction layer coming in, is there an equilibrium being reached between old and new? How would you describe what's going on? >> Fascinating question. What I often see in most of the enterprises I work in, is, they have a couple of investments going on. They're on a journey, a dev transformation journey, and a lot of that is, you know, really at the core of it, embracing DevOps. But what you'll see is, there's groups really pushing the envelope in these teams with cloud-native, micro service development, really all about speed, how quickly can they take small teams, get the idea into market? But then what you also see going on is, large sets of very valuable assets that data transactional systems, and how do you start embracing more, and more automation, to really reduce the cycle times, improve the service levels, and to effectively, start taking cost out of that full equation, that full life-cycle. So, what you're seeing, is a lot of automation coming into the existing IT environment. You're seeing a lot more of taking down of the silos of ops, and development teams, and that's going on in the core areas, and in the more cloud-native area, you're seeing, there's actually a common team put together, and they basically own the whole spectrum. They build it, they run it, the whole piece. >> You would think the competitive implications of this are huge. Without naming names, are you, at this point, able to discern patterns where organizations that are implementing this type of approach, are becoming more competitive, becoming more profitable, gaining share. Do we have enough evidence of that yet? >> Yes. Well, Gene, we were talking about Gene Kim earlier, and you can see, from a lot of the studies he has, that you'll see how much more effective and high-performance you're getting out of teams that are really embracing the best practices DevOps, and it is translating into financial results. So, you are seeing that bridge occur, but, part of what got me thinking about, is, what we were talking about earlier, the analytics that we've been exploring in the, in the team insights, and how the patterns you see, in how teams are interacting, and their code, and, you know, where are the core committers, the extended community, and extended community, the extended ring outside of that. You can begin to see patterns that are working well, patterns that are starting to have problems. It might actually be an architecture issue-- >> A self-healing concept too, if you think about it. This is actually taking it to like, social media has the same problem, on Twitter, runs with the same voice. You could have a zillion followers, and not have any influence, or have, you know, 100 followers and have a lot of influence, based on, that's no measure for that. You're getting at something that's more scoring-oriented, and analytical. That's interesting to me, I'm going to follow up on that, maybe another time. The question I want to ask you, 'cause I want to, I can't get it out of my mind, 'cause you mentioned the cloud-native, it's got me, kind of really, you know, riffing on this. We believe it's a multi-cloud world, right? And there's going to be a variety of clouds, not a winner take all, and they're all going to have differentiation, but having the traverse clouds is going to be really, really important. So, Kubernetes is kind of interesting to me, because you're looking at Kubernetes really kind of coming in and saying, hey, we could actually be a factor in orchestrating, and managing the sets of containers and micro services. And so, it's almost like a whole 'nother land-grab is going on around Kubernetes, because, it's so delicate. Can you share thoughts on that? Because, it's kind of nuanced, Kubernetes is, has got great traction in containers and micro services, but it's super-important. Why is it important, and why is it fragile, or is it fragile? In the sense of its importance, and not to be forked or tweaked. >> First, it's growing very rapidly. The use of containers for development and building, largely cloud-native micro service applications, is growing at a very rapid rate. And then, the ability to set-up these Kube clusters in different clouds, to be able to take advantage of the characteristics or services that are, are in those different clouds, including, you know, maybe you want to set-up a cluster near where your data is so you can have the processing local to that data, maybe you want to set-up clusters around certain security, or privacy, or regulatory policies. So, Kube is really providing, almost a platform-like layer for the containers, that is very robust. I wouldn't say it's fragile, but, with that flexibility, to setting that up, and where you want to setup that-- >> It allows customers to really figure out where to put workloads that matter. So, IoT would be a great use case for this. IoT, say, hey, you know what? This cloud is awesome at this and, put that app over there, and this one goes over here, 'cause it's got something over there that I like, but now, you need to have, I mean, is that kind of where, this is like, interoperability of networking in like, the 80s, in 90s, when that whole trend started booming is really its importance. >> Yes, yes-- >> Its openness. >> Well, the openness is critical. A lot of what we saw in distributed computing and the connectivity between clusters will be critical, but I do want to get to that point you mentioned on the openness; to me, openness is critical from a number of dimensions. One, certainly for inter-operability, and portability, but probably the most important is the rallying point for innovation, that you get these ecosystems, and with open technologies, which really is an open governance with open standards, you find a lot of creativity and innovation occurring within that base, and that, to me, is what really causes these environments to explode and take off. >> And if they can take that openness into the data level, then you're going to have a perfect storm of innovation, because now, you've got open source, which is thriving, and continues to be great, tier one by the way. >> And you're choosing to invest so much, and give back so much to the community. Not everybody does that, but you've made a business case for that. Why that strategy? I mean, it's IBM, you would think, you know, historically, IBM, very closed. But, you are almost overly-aggressive about your open source investments. >> Yeah. Not even sure it's historical, it dates back a long time, quite a while ago-- >> Yeah, that's true. >> Dave: You can go back, all the way to Linux. >> Yeah, Linux was the, they were the main player in Linux. >> You go back, obviously, the internet itself, TCP/IP, Linux, Java, Eclipse-- >> Track record's amazing. >> To me, all these industry breakthroughs, things that shape the industry are often, at its core, there were, at critical places, there was an open ecosystem, an open governance, open technology that really enabled it to just expand and grow at a tremendous rate. >> I think blockchain is perfect for you guys right now. A great example of, and in, might, people might be saying, oh, a little bit early, I think that bet is going to be playing out well. If you take the open source, and this whole digital value thing, very interesting. Well, I mean, final thought: what are you excited about right now? I mean, as an IBM Fellow, you get the canvas within the tech space, obviously, a lot to pull, it's kind of intoxicating these days. We kind of went down memory lane with some old ways, but, there's a ton of great new things happening. What are you excited about? I mean, what's getting you buzzed up about the current tech scene? >> The things that are really, I find fascinating and exciting now, is the different ways we're learning to apply AI, cognitive machine-learning into the different systems. We just, sort of, covering it just a little bit, in the DevOps space itself, but we're learning to apply it from the end of test, to understanding how we can predict where we have problematic code files, and how you would improve your test or skills, to the other spectrum of how is the community actually operating? Is the community healthy, is it growing? How are my projects and my teams working together? How healthy is that, are there issues that I have to start looking at? Do I have a design issue, an architecture issue, a squad issue? So, I can start doing that. This is all, we're learning how to take in big data and apply machine learning to this to get these types of insights. And to me, you know, that's just one spectrum of how we're applying it, but that's, to me, what's so exciting, is how we're applying it. You know, some of the examples that were shown with blockchain and cognitive, and in IoT, and AI. >> Dave is changing the game. The algorithms are coming out as more like libraries, not as custom stuff, and you've got the compute over the top. It's like, I wish I was 15 again, you know? What a great time to be in the tech industries, a computer scientist or any kind of science field right now. >> It is a great time. >> It's just a super time. Appreciate it, Dave, thanks for coming on theCUBE. Dave Lindquist, IBM Fellow, vice president of DevOps and the cloud at IBM, sharing his insight, great job. IBM's coverage continues here at day two, here on theCUBE, I'm John Furrier with Dave Vellante. Stay with us for our wrap after this short break. (percussive tones)