 Live from Las Vegas, expecting the signal from the noise. It's theCUBE, covering InterConnect 2016, brought to you by IBM. Okay, welcome back everyone. We are here live in Las Vegas for exclusive coverage of IBM InterConnect 2016. This is Silicon Angles, the CUBE, our flagship program. We go out to the events and extract the signal from the noise. I'm John Furrier. Our next guest is Moe Abdul, who's the VP of cloud architecture and solutions on the Blue Mix team now on a new role, going out there and doing all the architecture. He's one in the brain trust of Blue Mix. Welcome to theCUBE. Thank you so much, John. Great to be here. Thanks for spending this time to come on. I know we're late in day three of wall-to-wall coverage. Three great days with all the execs on. Now we get to get down and dirty on Blue Mix. So Blue Mix, obviously doing well. Gave him props on day one. Still got the naysayers out there. Good luck with that against Amazon. So you're always going to have that Amazon comparison, right? So that's out there. But what a progress from Pulse two years ago. A lot of progress. Tons of progress. You guys have been beavering away, running like the wind. Now you're hardening. What's the status? Where have you guys really knocked the ball out of the park? Where have you solidly put the foundation down and what areas are you hardening? So as you said, John, great progress. A lot of the hardening that happened was really based on curation of client experiences. Initially, of course, people think of cloud just as a public cloud. But as you heard us talk about and say a lot, cloud is for us in a hybrid setting. Really nailing that down, the ability for people to start to build applications and then consume them in the way they want, whether it's in a local, dedicated or public fashion, has been a lot of the effort last year. And of course, a lot of it is based on grounding, common standards and technologies that really lend themselves to that environment. The second area that I would say we're really hardened past the local and so on is defining repeatable solutions. A lot of times, clients have been doing first of a kind in the cloud and now we started to understand the common patterns around mobile or data. We're aggressively going on saying, we think you really should think about doing it this way. So obviously, cloud is running, and Watson's been sprinkled in there. So Watson, as Sandy Carter says, is the catnip. Yes. Okay, so that's obviously to give some differentiation. There's a little IBM twist to it, obviously, for differentiation. Microservices, we had Maria on earlier, Maria Week. She was talking about that. How is the glue layer working with Bluemix right now? Because obviously, the APIs are key. So everything is a service, is the concept. What's going on in the glue layer and all this? Oh, absolutely, great concept. So one of the things we worked on last year to your point about hardening is creating essentially a common operating system for the platform so that all of the services that are coming on board can easily connect. That operating system underneath it really has multiple runtimes. OpenStack being a key element for extracting the infrastructure. That's a key glue element. The second is around containers. As people start to build services, so for example, Watson is built on top of containers, which in turn leverage the same sort of common operating system services. So Watson can plug in via the container? Via the container, absolutely. So OpenStack Summit's coming up in Austin. Is OpenStack doing well? I mean, you guys happy with the OpenStack piece? Any of your thoughts on that? OpenStack is growing. I mean, at these days with these open source projects, it's like a race as to who's growing fastest, but they're all like sort of... OpenStack is healthy. Very healthy, very, very healthy. And you're happy with the progress of OpenStack? Absolutely. We've clocked in now. Probably we've exceeded the 150,000 contribution, something we haven't seen before in terms of people engagement and access. And so talk about the relationship between Cloud Foundry and IBM. Obviously you use Cloud Foundry. How does that relate into it? You guys, do you tweak it a little bit and obviously, and Pivotal is, I think it's part of Cloud Foundry? Yes. How does this all work together? So actually, this is a very, very important design point for us. Many clouds are built based on a single runtime, either something like a Foundry-based, which gives you that rapid application. Container-based, which gives you a lot of this sort of microservices. Or OpenStack, which gives you a lot of control in terms of how you want to configure applications. The reality is, when you look at what customers are doing, is that they're combining these things together. So Foundry for us, both of course, is a critical element for these mobile apps, rapid development. But it has to find a way to fit in. So when you think about Watson services, mobile services manifesting themselves, how am I going to be able to connect that back to some container application? So the work we're doing with them is much more around, you know, we've moved from just hardening the base to how do we integrate these ecosystems together. Okay, so standards is a big problem right now. People, customers want standards because if the vendors are in there, polluting, I shouldn't say polluting, polluting these standards efforts by having different versions, that is not good for the customer, right? You know, whether we had the IETF in the old days, the W3C, IoT in particular, big standards, kind of thing that's going on now. So that's a big deal. Containers right now is a big element with standards. Are people standardizing on Docker containers? Does it still jump ball at this point? What is the container situation? Today in the container world, we have really two very thriving and active communities. Docker, no question, is taken developers by storm. And of course, as people mature past the single container and they're deploying more complex applications, they're continuously adopting that ecosystem. Of course, containers existed for some time before Docker and others. And Google has another very thriving community in the form of Kubernetes. What's very interesting is that a lot of the folks who are using one or the other are active in both communities, which is where, of course, things like the Cloud Native Foundation comes into play. How do we actually start to solve common problems that exist, security? It exists in Kubernetes and exists in Docker. Can we actually work together to solve it one way so that whoever adopts one style of technology versus the other, they can still benefit? But they're both very active, very healthy, growing. And from an IBM perspective, we support both of them. Awesome, so let's talk about Kubernetes. Does that have traction or is it still exploratory? Obviously Google's got to put their weight behind it. And we saw the big news of Spotify moving all their infrastructure to Google Cloud. And that's a big deal, that's a big trend. Is Kubernetes making it happen, or is it what's your take on it? So what Kubernetes did, which was very unique in terms of how they came at solving the problem, is that they solved the issue of orchestrating complex applications. They came at it from that sort of angle. And so the whole operations management is much easier when you think about Kubernetes. What Docker did is they came, of course, at it from the developer productivity. Very easy for me to get started in a laptop and move on and so forth. Kubernetes has been really killing it in terms of the whole management, which is why- Or more orchestration. And the orchestration, absolutely. Orchestration seems to be a value proposition. Absolutely, orchestration piece. So what you now start to see from, let's say, an IBM container perspective is how do we marry both the orchestration layer but with a Docker infrastructure or a Docker API? So you guys are doing a lot of work. And Meg Swanson, when she was just director of marketing, now has done a great job. Now she's got a new role. Adam now running product management. Adam Gunther. The team is growing. You had a lot of good wins in the logo slide in terms of customer references, expanding rapidly. So you got some successes. Congratulations. Now you got to go out in the real world. Mainstream enterprises. You got the early adopters in there. So now the architect that's out there, they're under a lot of pressure, not to produce the five-year plan, but the one-year plan. And DevOps now is a horizontal infrastructure. So the digital transformation is the end game, but it's usually maybe app-specific. But the challenges are also on the infrastructure, from software to file, all the way up to the edge. So what's the playbook? How do you guys, what does a customer do? Because right now no one's really laying out the blueprints. Where are the reference architectures? What's the playbook? Are you guys doing anything there? Is there a group within IBM that does this? Oh, absolutely. Oh, your group. My group. Thank you for the, for allowing me to talk about that. You're absolutely right. I want to actually say something interesting, because when I talk to the customers, you know, things come in cycles. And we had this whole movement where developers took over. You know, it was all about the developer and the IT guys went sideways. And nobody really talked enough about the enterprise architect. You know, that prominent role that also diminished. What I'm starting to see is that those guys- Rebirth, rebirth. Rebirth. Rebirth, absolutely. Yes, they're back, it's revoked. Especially in a hybrid context, because everybody has their existing, you know, infrastructure assets, et cetera, and they're starting to marry all of these complicated, you know, technologies and capabilities together. Earlier this week, as part of, you know, Interconnect, we announced the launch of what's called the Architecture Center. So the Architecture Center is a single portal you can go to and start to understand from an enterprise architecture point of view, how do you actually put solutions together? If you're thinking about an IoT solution, a connected car, what are all the capabilities, best practices, and sort of starter, you like assets that you would want to use? We also start to expand it into industries. So if you're in the retail industry and you're trying to do some sentiment analysis all the way to inventory, and then if you like fulfillment, what are the best practices in which you would connect these systems? And what we're seeing is many of our clients saying, thank you, I don't want to start with the atom. Well, we're doing, so we see a problem here and opportunity for you guys, because your customers, I mean, everyone's about to, you know, all the analysts cover the vendors and say, oh, this is the silos, and they charge for it. We have free research at Wikibon. Brian Gracie's our analyst. And one of the things he's doing that we're putting a lot of effort into that's free is actually talking about the digital builder. That's Kevin Egan's term, but we used to call it practitioner, doer, the guy, the architect whose job is to get it done. Now, technical people don't go into, you know, stack overflow and say, can someone answer me, because they're too proud. They don't want to look stupid. But they'll search like crazy. So we're putting out content and the problem is out there is that there's no data on how do I horizontally scale an infrastructure. Do you guys have any patterns that you've seen, successes, data you could share with the audience around, you know, where to start for the architect, what ways to low hanging fruit? Okay, I get BlueMakes, it's cloud. I understand the services. I understand the API application. I get the vision. I understand the digital transformation objective. What the hell do I do? What do I start? So what do you guys suggest? What do you see? What's the pattern? So first and foremost, one of the things that people talk about is the domain level. But when you break it down below, the first step for success that we've observed is understanding how do you take your existing assets and start to enable them into the cloud domain, integrate them. So everything you've heard this week around APIs, APIs integration is actually a very low hanging fruit. In fact, I'm shocked and surprised that this has not been something that has been accelerated further. Now, part of why it hasn't been accelerated is that people just think about APIs as a connection point, as opposed to ways in which you can manage, throttle, understand secure, and so on. And all of these elements that now we've matured in terms of technology. The second low hanging fruit in terms of performance and scale is techniques that people have established in terms of how they buffer their systems of record. You know, in the old days, we used to capacity- Explain that buffering systems of record. Yes, in the old days, people used to do capacity planning. They go, okay, we expect we're gonna get thousands of users to our backend, and so on and so forth. In the world where you're getting hit through social, mobile, partner channels, you know, that scale is unpredictable, and the spikes and so on are, frankly, nothing that a traditional systems of record can handle. So buffering techniques start to enable us to actually create caching in the case of data. So you use things like CouchDB, et cetera, in front of your own data that you don't need to move, but you just sort of essentially transition parts of it. You can actually start to create routing applications. One of the clients that we've worked with, and I was really impressed with this, they were determining for all of their free tiers of customers, they would, for example, route them to a certain class of service for their paying. They never want to miss an SLA, so they would, of course, prioritize that back into the actual committed transaction. So this buffering technique is where you start to create a cloud, an edge to the cloud. You're essentially starting to create logic that understands what are your business priorities, what are your SLAs, what are your policies, and how do you interact with that system. That's essentially a very important component that I think people have a lot of, you know, they miss it. And there's now technology, in memory, you get flash, you get software to do all that, and you can auto-scale it, all kinds of good stuff. Okay, so BlueMix, let's step back and look at what you guys are doing. What is your group's charter? From what I'm hearing, the hallways here is it's a group of dedicated folks going out there with top customers, going down and looking at, and getting down and dirty, rolling up your sleeves, describe some of the things you're working on. And what you're trying to do long term. So first I would love to refer to us as ninjas, you know, where the people are black belts who are really starting to understand and fine tune. When somebody says, how do I get started? There's a lot of it out there. But when somebody says, I did get started, how do I actually scale this? And how do I make it fit in my environment? That's when we come in. We land, we work with the client, we understand the type of application they're building. We bring a lot of those patterns. We make it work in the context of their own application, infrastructure, et cetera. And once we get it all the way up and running, we actually continue to hand hold the team in terms of skills transfers and the like. The important part is the feedback loop. We then capture that back, we generalize it and we make it available to the rest of the industry. And that's a piece that I think has lagged in terms of the whole cloud world. Awesome. A lot of it is unique to a single customer as opposed to something you can use. Mo, I love talking with you. We should spend more time with you. We should do a whole day with your team. Just from the architecture. We'll come down to you and Raleigh. We'll come down to Raleigh and do an innovation day and get a deep dive on this. Really super, super interested, the progress you're making. You guys are running fast now. You're going out in those ninjas and getting the market going really fast and taking the big customers to the next level. Final question for you though. Share with the audience who's watching, who's not at the event, what's the vibe here? What's it like here? What's the show vibe? What's happening? I have to say, I was talking to a colleague about this and when you walk the hallways, last year was a lot about energy, excitement about learning new concepts and so on. This year is a lot about, I feel like I can do it. I'm doing it. I want to share what I did, which was very different from I want to learn to I actually want to share a lot of the best practices that I've ever heard. People will get some stuff done. You guys shipped whisk. We shipped whisk. That was a huge accomplishment. Congratulations. I know that was kind of coming really fast down the pipeline there and more stuff coming. More stuff coming. And the other thing is we're running out of time to talk about all the great things. People are like, I asked them, what are you excited about? Like couldn't finish this. You can see and just like, okay, boom, boom, boom, boom, boom. He's like the end is on the keynote on day one. That's right. He needed a whole day. That's right. But you know what? I think, John, this is an important thing. Just, we are just like our customers. Speed is the new currency. Speed of innovation. Speed by which you're reacting to these requirements. That's what I believe. So I'm excited about that. We do too. We're speed. We've speed the Cube interviews. We have a new innovation called Cube Gems that are speeding up this interview. It's a highlight. Go to twitter.com. And look search on the Twitter feed, on the search button on Twitter, Cube Gems, hashtag Cube Gems. Mo, you're gonna be already up there. Highlights of it. You can get a little taste. And obviously get the full interview on siliconangle.tv. We are here in Las Vegas for exclusive coverage. This is the Cube. I'm John Furrier. Be right back after this short break.