 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. theCUBE's exclusive three-day coverage on day three, great interviews coming up. Our next interview is an IBM fellow, Shenkar Kalyana, IBM fellow CTO of the cloud group with IBM and GBS, Global Business Services. Welcome to theCUBE. Hello, welcome, thank you. So I'd love to have IBM fellows on because it's the status within IBM. You get a lot of range to go play and dig around and have fun. And you do it with customers. Bringing the technology piece in, really, really kind of super important that you guys have that range, but also gives you depth. You got like your helicopter. You can go high level and then go deep under the hood. So this is really interesting. You guys are doing a lot of cloud migration on stage. You guys are showing customers. You have proof points on some of these cloud migrations, not trivial, but not super hard either. I mean, I'm not trying to oversimplify it. What are you guys talking about on the stage around cloud migration? A lot of customer testimonials. What's the key migration message? So I think the first thing we should, what we're seeing is the use of the word migrate itself has got some connotational differences. The pure technical purists would look at that as, hey, it's about moving stuff from here to there. And that's all it takes in walls. That's pure migration in the pure sense of the word. But the customer view of the world is the landscape is that that move is hardly ever pure lift and shift. I got to do a lot of things around it. We talk in terms of the various stages of life cycle it can actually go through. You start to do nothing, which is like, well, you look at the workload, nothing gets done and it's okay, retain it as is. Or you have to discard the workload, archive it, consolidate it, deprecate it. Or you got to move it to a cloud because it runs in a better infrastructure, so to speak. Those are like the simple views of the world. Where it gets to be crazier is, well, guess what? When I did that, I need to remediate the platform just a bit. I got to change my Oracle 8.1 version to an Oracle 12.2 version just a bit. Not a big deal, but just a small thing. But you unpack the thing a bit more and you find out that, well, the database on the source side and what I got to do on the target side and the dependencies I have all over the place it just gets to be a bit more. And now you graduate from, well, it's not just platform remediation, I got to remediate my application. And then you say, well, I started remediating that. I found some dependencies, I need to go rewrite some pieces here. I got to re-architect stuff. And so it's a progression from do nothing all the way through rewrite and re-architect that this phase is where the migration problem is going to become. You know, John Granger's on stage talking to the customers and I had a moment when I was looking at that saying, you know, you mentioned lift and shift. I mean, I've heard the word rip and replace. These are words that have been kicked around IT. Rip and replace, lift and shift is now the cloud version of that. And some new cloud players will say, hey, just lift and shift or use my G Suite or my SaaS and that's now your cloud strategy. But it's harder than that on the enterprise. And the quote yesterday from Don Tapscott, said if God created the world in six days because he didn't have an installed base, right? So, making it, moving to the cloud with pre-existing conditions like workloads that are running the companies is not trivial. Take us through how that gets done. Because it's end-to-end now as with a vision. Take us through that methodology because the install base matters. Yeah, absolutely. I think in the journey we typically go through with clients and clients actually go through, right? And that's what we are codifying for them is whether you're talking about a single application or whether you're talking about a portfolio, you got to really understand the business drivers behind it in terms of why are you trying to do, what are your real objectives? Are you trying to create an innovation platform? Are you just trying to reduce cost at the run cost, so to speak? Are you looking to change your delivery the way the delivery itself occurs, what we call delivery transformation? What are the different objectives that really guide that? And examine the portfolio, go through a pretty, I would say, a very prescribed way to analyze a portfolio with tools, with methods, with discovery techniques that end up with dispositions, right? So you can say I got like 1,000 applications, 20 of them had to retire, 50 of them need to be consolidated to something else. I need to migrate because these 20 different workloads to your point may be very simple and I don't have to do anything about them, just move them to the cloud as is. But I got this other portfolio about, hey, you know what, I got to modernize, I got to remediate, I need to do that. So you end up with dispositions for each of these and each of them now have what we call execution patterns that take you along the journey, right? So in the pure lift and shift case that really nothing gets done except the pure lift and shift, you can do that like that, the way most of the people do and that's... It's okay for certain situations. Absolutely, those situations do exist. Where it gets complicated. That's the benefit of the cloud, the cloud innovation is you can actually use cloud for certain things that are great for the cloud, like analytics or... Absolutely, absolutely, right? But every one of these execution patterns has some different techniques and different tools and different assets and different things that come to play. Your capabilities become different. The guy that's doing legacy code conversion from Koval to Java is not the same guy that's actually doing systems of engagement application on native, right? So yes, in general it is cloud but the skills kind of start varying. So you need to be able to understand that the journey map of how this execution patterns actually go through and when you go through that path you'll find that, well, I got a... I just advised you on how to go through the journey map which is advised on cloud adoption. Now I'm helping you execute on whether it's migration, whether it's modernization, whether it's rationalization of the portfolio or if I'm creating new white space innovation and whether it's on Bluemix, whether it's on Amazon, whether it's on anything else, doesn't matter. So I'm interested in sort of exploring that a little bit because when you hear migration, okay, oh, migration, expensive, complicated, time consuming, risky and Ginny said yesterday, look, this cloud is not just a new way to deliver IT, it's a platform for the future. It's not just a better way to do things that you used to do. It's a way to deliver things that you've never done before. So now I'm sure there's certain applications you can lift and shift, great check. But what's the conversation like Shankar in terms of, well, I want to do something that's not just one little one-time cost savings. I want to do things that are going to transform my business and give me significant competitive advantage. What are those conversations like? Yeah, so this conversation was actually a continuum, right? Because one of the things that clients and our experiences have led us to this thing too, which is you can talk about this journey maps and you can talk about this story of month-long transformation to get to the level that they really need to. But what clients need is the proof of value that says, look, show me something now, get me something off the board quickly. So what you typically look for are those initial, easy to do workloads, proof points, that you can show the value, because a lot of this thing is value begets value, right? So when they see things working, they start getting belief in that because the other part of the conversation is, we use the word culture, but it's really, a lot of the people that we deal with, they haven't really used to this ecosystem of this complexity. They think cloud on one hand is extremely simple. It's like, I'll click and everything needs to start. It's auto-scale. It's auto-scale. I just click and it all automatically happens, right? In fact, we have this discussions with clients about in this entire journey, you can't forget the single architecture, right? I mean, it just doesn't happen with magic. You still have to consume services. Someone's got to put the system together for you, right? And you had to do that. You had architect, high availability. You got architect in a four line system. It didn't, it just didn't come magically, right? So the journey point is about describing that vision of what it takes to do, but you had to do that in increments and be able to take that journey with them so where they're actually learning alongside and experiencing that. So we usually advocate a proof of value as a very first step. Migrate three workloads, build a native application, do an MVP of something that can show value. AP enable a specific legacy functionality, et cetera, et cetera, right? So you have different kinds of these exemplars. Depending on scenario, you can actually do that. Okay, and there are a lot of workloads where you can get some quick hits and some wins. And despite all the talk of cloud and certainly public cloud and the ascendancy of public cloud, the core of enterprise IT apps has still not moved to the cloud. When you have conversations with your clients, should they move? Why should they move? What are people thinking? All of these are pretty valid questions in the sense that when we look at the portfolio analysis, we are applying pretty algorithmic and human intelligence based kinds of metaphors there to be able to look at a workload and say, what should be the disposition of the workload? Should it stay? In fact, there are workloads which should really require specialized systems and you want to keep them exactly where they are that may not be necessarily cloud-ready. And that's totally okay. In fact, that's internal advocate. You don't want to paint the town with one color and say that's the way it is. On the other hand, there are workloads you could say, well, I want to take advantage of these drivers on the cloud, but I want to retain a resident whether it's data residency requirements, security requirements, whether it's just the initial culture structure I don't want to move it there. So you kind of get them along the journey to say, well, maybe we want to help you create your own cloud and then start from there. In other cases, there are certain distinct advantages the public cloud actually provides in terms of its innovation potential, the way the services can connect to each other, what kind of service portfolios are actually available to you. So I think in some of these cases, clients actually look at the promise of that. And that's where the proof of value to me is very important. Well, that's the thing that we're talking about now is that the innovation upside for the cloud and the cloud promise and how the hybrid cloud is coming together with the surge of cloud native with microservices and containers and Kubernetes is showing a lot of great promise. People are super excited about that. And hybrid is just the workhorse. It's the brute force blocking and tackling. So I want you to take a minute to explain how you guys do this because this is the needle you're threading. Explain how you guys are bringing the capabilities of GBS and GTS, two different groups with an IBM, to make this happen. And how does that kind of pan out on the course of your engagements? So let me set the context in terms of, there are the solution view that I typically look at and we typically look at is what I call the two by two. So you got this lifecycle view that's the journey from how do you go from idea to scale? How do you go from advice to operate? And I'll talk about operating a bit. The other dimension is the solution stack, right? Which is the business process, the cognitive business process, the application logic, and the data logic that powers it and then the platform that actually enables it, right? So in this two by two, that every solution as a journey goes through, clients typically have been placing, if you look at the two by two as a cell, a bunch of cells, people have been operating in different cells, right? So the clients will look at that and say, well, you know what, I'm going to have, capability X for that one cell and vendor Y for this other cell and someone else to put this all together for me. So that's really what's been happening in these. And they think they're all good because all the boxes are checked. Wait a minute, we've got a blind spot. Exactly. So that's what you're getting at. The technology stack and the solution stack, sometimes people get misconfused between the two. They're complementary. Exactly. And so where we, when the power of IBM services is really being able to bring the full power of this entire stack and this life cycle together in one unified way, right? And that's what Cloud Innovate that the method actually does. Because what we've really done is map the journey pretty well based on the demand profiles. Now, every client doesn't need to go through all of these things. Their entry points could be different. We have clients who have come and told us, look, I know exactly what I want to do, help me to get there. There are clients who want to take the whole journey with you, right? So it sounds great. Can you give me some examples, proof points? Yeah. Okay, I'm interested, what can you tell me about that? Yeah, so let me talk about a couple of different examples, right? So one is Etihad that they were on a stage with us yesterday, Etihad is a global airline as we know. And one of the things that we have done for them, they are proof on one of the pain points, amongst the many that they've actually gone through. One of the classic things that they had to do was to take their enterprise ERP systems that have been running their business. In fact, they are about a 13 year old company in the annals of IT, that's like nothing, but really it is legacy already in this modern age. So their journey was how do you actually create their service business for the future by taking advantage of the Cloud? So they had a problem of how do you actually take the ERP portfolio and take advantage of the Cloud? So we helped them with joining to work with them on the SAP applications that they had migrating to the Cloud, Diabium Cloud. It's a private Cloud, that got set up. The other example, another spectrum is American Airlines, right? Their journey is about way much longer, but for them it's an enterprise transformation that is actually in progress and they want to be able to get to a completely different point of Cloud becoming a true innovation platform. So they want to be pure fully in the past as a vision, but they understand that they got to go through the IaaS. So I kind of talked about the sexiest way in Cloud is, oh, it's all cool clusters, past base. In fact, forget even Cloud, let me go SaaS and everything works beautifully. That's ideal Nirvana. I think if we can get there, that's a day a lot of the time. If you have no legacy, if you have no stock base. But the reality is you're going through a granularity of some workloads on bare metal, some workloads on VM, where some that I had to containerize and et cetera, et cetera. So in the American Airlines case, we have had both the journeys, right? So we have a journey about how do you rewrite, rather than the rip into place analogy that you did, how do you rewrite portions of it completely on the native Cloud while you take advantage of the IaaS infrastructure in front of you to be able to gain some cost efficiencies, resiliency efficiencies, et cetera, right? So you got these scenarios, you know, playing out in all these lines globally, right? So. Well, I think you guys got it right. I mean, looking at how you're doing the migration, I think you got a good approach of taking the GTS and the GBS together. So I think what you were bringing in is that, the entry point is different because every company is different. There's no more general purpose software anymore. There's no more general purpose infrastructure. It's specialized, but standardized at a lower level. Yeah, it is tailored, right? And the composability of Cloud Native gives them that vision for the future, but why not go Cloud Native if you can? But bringing the technology stack and the business stack together, I think really is fundamental, I think. And that's what you guys are doing. Absolutely, absolutely. And I think the one of the other messages we try to do in this is, you know, there is very acceptable level of impatience in front of clients, right? I mean, they don't want to take this journey over several years. They want to still get, I always look at migration is still a means to the end. It is not the end. Clients are not there looking to migrate. They want to get to the cloud and start innovating, right? So the faster you can get the migration journey done, is better off you are. So what you don't apply is just people and expertise. What you apply are tools and innovation and automation to be able to get to this point. So a lot of the work that we are actually doing is in getting a better understanding of, you know, how quickly can you understand and analyze the source systems and arrive at those dispositions we talked about and then be able to create the target environment much faster so you can start migrating. So it's a total end to end. Compressing the time frame is so important, right? So, yeah. Well, Shankar, thanks so much for coming on theCUBE. IBM Fellow, great perspective. And thanks for the insight. Thank you very much. Thanks for hosting me here. Pleasure, all right. And to end, multiple stacks dealing with the cloud migration story here at IBM, great success points. This is theCUBE bringing you more coverage from day three, exclusive coverage. Stay with us. We've got some great interviews coming up. Blockchain, cloud, big data, IoT all coming up next. Stay with us. I'm John Furrier with Dave Vellante. We'll be right back.