>> Narrator: Live, from Las Vegas! It's the Cube, covering IBM Think 2018. Brought to you by IBM. >> We're back at IBM Think 2018. This is day three of our wall to wall coverage. My name is Dave Vellante and you're watching the Cube, the leader in live tech coverage. A lot of times in the Cube, we talk about how CIO's understood a while ago, they just can't take their business and put it up into the cloud. Rather, they have to bring the cloud operating model to their data. So that's a topic that we're going to talk about with Dave Linquist, who's here. He's an IBM fellow and Vice President of Private Cloud at IBM and Ajay Apte, who's a Distinguished Engineer of IBM Cloud Private. Gentleman, welcome to the Cube. Good to see you again! >> Good to see you Dave. >> Thank you. >> So, Dave, let's start with you. IBM Cloud Private, you heard my little narrative at the beginning. I think it's consistent with what your philosophy is, but what is IBM Cloud Private? What's it all about? >> Sure. Well why don't we just start with, there's public clouds, private clouds, hybrid clouds and the ability to match your workload requirements with the particular cloud, is very important. And having that consistency between private and public, so you have that flexibility, whether it's security, performance, cross aspects, regulatory, et cetera, is an important part of a multi-cloud strategy. With Private Cloud, in particular, we introduce Private Cloud, the offering is called IBM Cloud Private, last year. And the demand has been through the roof at the enterprises. What we're effectively doing, is bringing cloud-native technologies, right into the enterprise. It's really quite cool. We're bringing Kubernetes and containers into the enterprise, optimizing a lot of the core enterprise middleware, so it runs on this optimized Kubernetes environment and then integrating it with the security and operational systems of the enterprise. >> So as you said, you only recently, really, announced the IBM Cloud Private and you talked about private cloud for years, as did others. But others, maybe, had an offering, but the offering really didn't work. It really wasn't the cloud experience, so what did you guys have to go through... I mean, it's not trivial to get that cloud experience. So maybe Ajay, you can talk about, sort of, how you got there and what you had to do to get there. >> Right. We started with some use cases that we had in mind. So let me talk about three, very core use cases that we started with. The first one is, IBM has an anonymous enterprise grade, production ready, footprint of middleware in our customer's data center. We wanted to bring that footprint to a containerized wall, to a cloud-based operational model. When I say enterprise grade footprint that customers have today, they measure the success of that footprint in terms of KPIs, in terms of resilience, in terms of reliability, in terms of security and compliance, these kind of things. We wanted to bring the same qualities of services to a private cloud model, in a container model That was probably one of the main use cases that we started targeting. On the other side of the spectrum, the cloud-native micro-services based department. This is where most of the developers are interested in today. This is where really high velocity, agility, can be achieved. So that was the second use case that we were targeting. In both those cases, the key also is that customers already have existing tools and practices, those kinds of things, the data center. The idea was to very seamlessly integrate into that set of tools and practices and even people within the data center, while providing the same cloud operational model. And then the third main use case was around integration. By integration, there are various dimensions to integration. There's integration between the footprint that's running on PrIM with the things that are not running in containers. They my be running in DMs or bare metal instances or maybe whole systems running on our main frame, like IBM Z systems, right? And then there will be other services, may be running SAS services in public cloud, so the integration scenario is basically expanded from our legacy footprint all the way into the public cloud SAS connector, so that integration was the third use case for us. So those three use cases, I would say, became the foundation of what we did over last one year. >> So Dave, in thinking about, you know, bringing the cloud-operating model to the data, what should clients expect, in terms of that experience? Is it substantially similar? Identical? Are there huge gaps? What do you tell people? >> Well, that's a good question. What they're going to experience is, when you're using public cloud environments, what you'll see is your developers get rapid access to the content they need to start developing applications. And it fits very well into their agile DevOps life cycles, high iterations. And what you'll see is, operations teams often refer to it as site reliability engineering in a cloud model. They have access to all the efficiencies of cloud for deployment, scale, recovery, maintenance, all those types of pieces. So what a customer will experience is we're bringing those capabilities into the data center, but as Ajay pointed out, we're then able to run a lot of the core transactional data, analytic, messaging workloads right on that environment, so the developers get rapid access to that type of content, what they need. And the operations, can leverage those capabilities on a cloud infrastructure. That's the experience they're going to get, matching up the enterprise requirements with the cloud-native. >> Is the impetus to take that proprietary data, that 80% of data Ginni Rometty talked about that isn't searchable on the public web. Is the impetus to get leverage out of that data, that they don't want to put into the public cloud, or is to modernize their applications and cut their costs? Probably both, but I wonder if you can talk to-- >> There are many higher level, type of scenarios and use cases, so one that Ajay went through is, really modernizing your applications, extending with innovation. But as Ginni talked about, and I think, you probably had sessions earlier on IBM Cloud Private for data, what we're seeing is how we can bring many of the critical data services together, from data science experience and data analytics and data governance and movement and management, into this cloud technology, so that it can be used against the data that's in the data center, within the enterprise to start getting insights into that data and furthering their business. >> Ajay, I wonder if you can take us inside the development process, even the thought process behind how you approach this. The secret sauce, how you approach this challenge, maybe, differently, than historically, you've approached system design? >> Right, so since the whole idea of IBM Cloud Private is around cloud operational model, high velocity, agility, those are the things we are preaching to our customers. The very key principle there is, using those in our development, as well. Our development itself, is built on the same, open source DevOps tool chains, the cloud operational principles, so that we can achieve the exact same velocity, agility, that our customers are expecting to achieve with the kind of offerings that we are trying to make over here. So that's, sort of, the first key principle for us. The second principle, is around production readiness. When we are expecting a customer to run production-ready workloads, we have security, compliance, reliability, these kinds of things, the same principles apply back to the platform that they're going to use for running those workloads, as well. So the first thing is, we are our own customers. We have to apply the same principles to our platform, so that customers can do the same thing. Our platform is, sort of, a layered model, where we have Kubernetes and Cloud Foundry as the containerization model, but we also have a plethora of IBM and non-IBM and open source middleware software, that's running on top of that. And then, we have customer applications running on top of that, so we have to make sure that as we build this platform, all these layers are taken care of, in terms of how we can deliver a production-grade offering end to end. Like, when we talk about Watson Studio, what Ginni mentioned yesterday, running as part of ICP for data, for example, The idea of running that, where it's not just about ICP running a database, it's about what happens to the life cycle of the data and how ICP gets designed to make sure the life cycle of that data can be managed in a containerized model. Those are the kinds of things that became very important for our philosophy. >> Having a little fun, our development team rocks! They are incredible. What our organization has done, it's fully embraced all the agile DevOps capabilities, it's all developed on a cloud environment, we actually use ICP in our development of our IBM Cloud Private. It's weekly iterations, two week sprints, and every quarter, we have a major release. We've done that the last four quarters, we've had a major release come out. It's really been exciting. >> So one of the great things about shows like this, is that you can't walk around without bumping into a customer. So, my question, Dave, is what are they telling you? What's resonating with the customers, in terms of the services that they're consuming? What are they like? What do they want? What are they asking you for? >> So we did what we consider a soft launch in June, where wanted to get some experience and feedback from users and operations. And what we actually did, is opened a open-select channel with our users. So we had tens of thousands of downloads that came with that very first release and we got feedback continually on what they liked from content, how they liked the environment, the whole experience. In the beginning of the fourth quarter, we did a major launch with all the middleware capabilities, that content on the platform, it just took off. Since that time, we have upwards of 150 global accounts picked up IBM Cloud Private and started and going through the deployment, some are even going into production. The thing that resonates with them so quickly, is they have so many existing workloads that they've been trying, to really, bring into this dev transformation, trying to bring into cloud technologies and this creates a journey, a path for them through application modernization and then adding all kinds of innovation with micro-services for refactoring or even adding Watson Artificial Intelligence Services into the environment. >> Ajay, I started off asking you, sort of, where you got the motivation, a good starting point, your answer was outside in. You started with the customers, looked at use cases. Having said that, you're trying to replicate, mimic, to the greatest degree possible, the public cloud experience, so there's a reference model there. So when you think about what's next, do you, sort of, pop over to your public cloud colleagues in the IBM Cloud and have a little bake off and see? Where do you get your motivation going forward, your, sort of, road map ideas. Obviously, the customers, but do you benchmark yourself against public cloud to try to close that gap? How do you approach that? >> Sure, there are multiple dimension. Customers, of course, is one of the important ones. Having a consistent story between IBM Public Cloud and IBM Cloud Private, is an absolute key principle for us. It's not just a requirement, but it's not just about keeping them functionally consistent, keeping them expedience-wise consistent, but making sure that when customers embark on the journey of hybrid deployment, be it, in terms of doing my dev test in public and then moving to IBM Cloud Private for production, or be a bursting scenario, these kind of things. Customers, not only want to run their application seamlessly, they want performance, they want network connectivity, they want secure connectivity, these kind of things. So that becomes another angle, in terms of how we are growing this, we have public, we have private, we can build a seamless hybrid storage today, but how do we evolve that hybrid storage to make sure that we can give them the same qualities of service? Just because you move your application from private to public, if the data stays on private, the performance is going to really impact, it'll suffer. How do you make sure that those kinds of things are taken care of when customers truly build that? So that's the second dimension of how do we really take the customers on the hybrid journey? And the third important one, is that customers, of course, are going to deploy on our cloud, on other clouds, right? They will always have multiple clusters, geographically distributed. How do we manage their entire footprint and give them the right views for deployment, management, accountability, these kinds of things, across that entire real estate, right? What we generally call hybrid cloud management, multi-cloud management. >> And that's a really, fundamental technical challenge, presumably. To create that similar capability, that consistency, maintaining performance. You've got a got of challenges there. Good thing these guys are rock stars! Alright, Dave. We'll give you the last word. If you had to summarize Think 2018 in less than 10 words, what would you say? >> Accelerate your transformation with cloud. That's what I would say. Leverage the technologies across IOT, public, private cloud, AI, block chain, and accelerate the transformation. >> Ajay, Dave, thanks very much for coming to the Cube. Good to see you again. >> Thank you. >> Alright, keep it right there, buddy. We'll be right back with our next guest. You're watching the Cube, we're live from Think 2018. (techno music)