 Live from San Francisco, it's theCUBE. Covering IBM Think 2019, brought to you by IBM. Welcome back to San Francisco everybody. You're watching theCUBE, the leader in live tech coverage. My name is Dave Vellante and I'm with my co-host, Stu Miniman, Lisa Martin is also here. John Furrier will be up tomorrow. This is day one of IBM Think. Kind of the pre-game, Stu. The festivities kick off tomorrow. They're building out the solution center. They got Howard Street, Takeover, we're in Moscone North, stop by and see us. Daniel Berg is here. He's a distinguished engineer with IBM Cloud Kubernetes service at IBM of course, Dan. Thank you. Great to see you again. Thank you very much. So everybody's got a Kubernetes story these days. What's IBM's Kubernetes story? So IBM has taken a big bet on Kubernetes two, two and a half years ago. Never really looked back. It's our primary foundation for our platform services. And we have two key distributions for the Kubernetes service. We have IBM Cloud Private, which is a software distribution for on-premises, set up your own private cloud based on Kubernetes, behind your firewall. And then we have a managed service in the public cloud. So you're moving to public cloud, doing cloud native, grab an API, CLI, you get a cluster. So a lot of people think Kubernetes is all, I can be able to move it anywhere, private cloud, public cloud. But there are other benefits of just, say for instance, a private cloud. Maybe explain those. Yeah, I mean, the biggest benefit for us is that we're able to give you the IBM Cloud experience and IBM Cloud content. So IBM content, middleware, things that you've been using for a decade. We've modernized it, put it in containers, install it and manage it on Kubernetes. And the nice thing is that content you can bring on-premises where it's needed the most and run it in ICP, IBM Cloud Private, and also take that and run it in our public cloud. As you migrate and move those workloads into the public sector. Dan, one of the things we've been watching is you talk about a hybrid cloud or a multi-cloud world. There's a lot of pieces and it can be complicated. Now, Kubernetes itself, not exactly the simplest solution out there, but when you can deliver it as a service, you can take a certain piece of your environment and IBM helps to simplify that. Maybe explain what it simplifies and what still are some of the hard places that we have to play at in these environments. Definitely, so, I mean, the IBM Cloud Kubernetes Service, we, anyone that has dealt with Kubernetes knows it's easy to install, pretty easy to set up, and basically easy to get started. It's the day two, it's the operations, it's the long pull, it's doing all the updates, the maintenance, the security patches, the securing it, making it highly available. That's hard, and that's hard over time and it takes a lot of resources. So, IKS is a service that we do that. Let the experts do it is basically what we tell people. We are experts at managing Kubernetes. We do this as our day job, 24-7, right? Literally, because we manage a 24-7 service. So, we operate at 24-7 and we keep it updated. That allows our customers to focus on their business problem, focus on their app, not building the platform. But there are still some complexities because you don't have just one cluster. If you only had one cluster, it'd be no big deal. I probably wouldn't have a job. But you have many clusters. You've got development clusters, you've got test clusters. But if you're doing a global service, you've got many clusters throughout the world, highly available clusters. You put clusters in various data centers for keeping your data in one location, right? So, you've got many clusters. So, it gets complicated to manage all of those clusters. So, with Kubernetes service, we provide all the capabilities to manage and set up and secure your cluster. But then the content, like moving and configuring things across all those clusters becomes complicated. And that's where we released recently a new product called Multi-Cloud Manager. Tell us more. I thought you were going to ask a question. So, Multi-Cloud Manager, what it basically does is it provides a control plane that allows you to manage and today it manages resources, Kubernetes resources across many different clouds, across many different cloud platforms. So, it works with our cloud private, which runs on premises, but it also works with our public cloud, IKS. And it can work with other cloud providers. It can work with Amazon. It can work with Google. It can work with Azure. And it works with OpenShift as well, obviously. So, having that one tool then gives you the mechanism to drive consistency of the resources across all of those distributions of Kubernetes clusters that you have. And another big thing that it does, and it helps with, is security compliance. So, it has ability to define security postures that you need to have across your clusters and then apply it and run it in both a check mode to see is that policy enforced or provided across all your clusters and where do you have gaps? And then it also has a setting to do enforcement. So, if it's not there, it'll make it there. It'll make it so. So, IBM hides all that complexity from the customer. But I'm curious as to what the conversations are like, Dan, with the customer. In other words, you're basically figuring out how to do it. The customer knows what it's doing. Do you ever get into a situation? No, of course, at scale, you want consistency and standards. So, do you ever get into a situation where a customer says, well, I'd like you to do it this way, and what's that conversation like? Yeah, so that's where, and that's where it's nice having multiple distributions, right? So, in our public cloud with IKS, having variations and unique configurations for each and every customer, I don't, we don't do that, right? It's a service. And services scale and provide value by doing consistency, right? So, we consistently set up and manage clusters thousands of them, tens of thousands of clusters that way. But if you need something that's highly, highly specific to a given use case, or you have differences in your infrastructure that you need to have more flexibility, that's where our IBM Cloud Private comes in. And we do have customers like, especially on-premises, right? On-premises, those are unique beasts, right? The infrastructure, the hardware, the network, you got to have a custom configuration. So, coupling our ICP product with the global services team, they can come in and they can customize it to suit any customer's needs. So, Dan, you talked about living in multiple environments, whether that be public cloud, your private cloud, you also mentioned Red Hat, I think, in there. Tell us where customers are today with OpenShift, where that fits, and give us a little bit compare contrast as to what IBM's doing today. Yeah, definitely. So, and it's interesting watching what's happening in the industry, because there's the whole push to cloud, and everybody knows they want to get there. But trying to get there all in one fell swoop with all the workloads that you have on-premises is quite complicated and difficult and almost impossible to do on day one. So, the story is all about how do I modernize what I have today on-premises? And how does IBM help with that in my journey to move into public cloud? And that's where, I know it's a buzzword, but hybrid cloud comes in. But for me, the hybrid cloud and what our customers are saying is that, I want to modernize what I have. So, give me a platform there. And ICP, IBM Cloud Private, and OpenShift are the two best products in the market, bar none, that provide that experience there. And our ICP runs on top of OpenShift. So, for those customers that have already been invested in the OpenShift space, you still get the value of IBM's content and integrated monitoring, integrated logging right there in that product space on the platform where it's there already standardized. How do you define best? What are the attributes of high quality and best? So, I guess best is kind of difficult to really define. But for us, it's all about ensuring that we have a solid platform, a solid strategy and technology set that we're building our offerings from. And we gain a lot of experience from our public cloud because we built and standardized on Kubernetes. We provide Kubernetes service and we do that at scale and secure as well as highly available. We take a lot of those same lessons because we have hundreds of customers running on it at scale. We take those lessons and we help evolve our private cloud offering as well. So, we bring those down, we provide a very tuned, somewhat customizable, but highly tuned supporting IBM content in that environment. So, when I say best, it is definitely the best platform for running IBM content, right? It's tuned for running IBM content, bar none. Okay, and my other question is, you mentioned hybrid, you said it's a buzzword. Okay, fine, but at least we know what hybrid is. Yes. You got resources on-prem, you got resources in the public cloud. Multi-cloud is the other buzzword. Sometimes we worry that the companies that are vendors like yourselves going after this multi-cloud opportunity, which is clearly a large opportunity and one that's needed because I want a consistent way of managing at scale. But there seems to be a lot of different initiatives within organizations. There might be different lines of business, there might be international people. Are you seeing any hope or sense that the customer constituents are getting together, the different constituents saying, hey, this is the strategy that we want to use to manage all of our clouds, or is it sort of, you know, fiefdoms that are popping up? What do you see there? Yeah, so it's funny when you do go into a large organization, a large enterprise, you're having a conversation, they've made a choice down one path using, let's say, IKS as an example, but then you realize you're having another conversation with another group that hasn't made any choices. I don't think that within an organization, within a large enterprise, coming together and saying, we're all going to go down one path with one tool to rule them all. I just don't see it, right? And also, even just going down the path of saying, I'm only going to stick and use one cloud vendor, that's also somewhat a thing of the past. You don't see that anymore, or at least where customers are moving. So within an organization, yes, you still have the lines of businesses and they might have different tools and they might decide on different tools and how they manage their environments. But the thing that customers do need to look at and what they do need to standardize across an enterprise is just some of the core tenants and the core technologies. So for example, if they're moving to cloud, whether it's on-premises or off-premises, what's the platform that you're going to build to so you have portability? It's got to be Kubernetes, right? That is a decision that as an organization, as an enterprise, you've got to agree on as you move forward. Because whether you use the same provider or the same set of tools doesn't matter as much, it'd be nice, but you've got to have some agreement on the core technologies and platforms. Because ultimately you can get there. It might be a little harder, but still if your core Kubernetes, it's going to be easier than different flavors of Unix, for example. There's at least a path that as they mature and as they simplify and they converge, they can do that seamlessly. Back to the cloud monitoring tool that IBM has. Who's the constituency? Who uses that? And give us a little bit of color inside, kind of the administrator, developer, cloud architect, what are you seeing? Yeah, so that's a great one. The cloud monitoring, IBM cloud monitoring provides visibility into your workloads within your environment. And that's not specific to just Kubernetes either, right? There's Kubernetes, but then there's VMs and bare metal workloads, more traditional workloads, that the monitoring service works just fine. Our developers have to have a monitoring solution. You can't build a cloud-native solution without monitoring, right? Monitoring and logging, it's like peanut butter and jelly. You got to have them. And if you're building a cloud-native solution, you're building Kubernetes, you're dealing with multiple clusters. Like I said earlier, hundreds if not thousands of workloads, you can't log into each one of them. You need a system where you can monitor and log. So the monitoring service is necessary here for simple developers to understand what's happening in their environment. And our partnership with Cystig provides us with a very rich monitoring solution, which we've done extensive integration in IBM cloud to make it simple for even developers. They don't have to go and install and set up Cystig themselves. All they do is, is a simple, I want a new instance. Directly in the IBM cloud catalog, they get an instance of Cystig and it gets installed into their cluster and they're off and running. It's simple to have. And we're talking visibility on things like performance management, security, problem, change management. Yes, yes, absolutely. So you get, and obviously that's all configurable, but what's nice with Cystig, and one of the reasons I like it, especially as a developer, as soon as you turn it on for one of your clusters, there's so much rich data that's available there, just out of the box. And they support other projects too and provide integration, deep integration, like the Istio project, for example. Great, great little project for service mesh. Cystig supports that out of the box as well. Built in, pulling metrics, dashboards built specifically for Istio. And I don't have to do anything as a developer. I just turn it on and then I start watching and seeing all the metrics coming. So it's kind of day zero here at IBM Think. Dan, what are some of the things that you're hoping to accomplish this week? I know you got a bunch of customer meetings, some of the things you're excited about. Yeah, definitely lots of sessions, great sessions, but it is the customer meetings I'm most excited about. I have a large number of them. I want to hear what they're doing, right? I want to understand a little bit better what they would like us to do in moving forward. How can we help them? How can we help accelerate their adoption of cloud, getting on the cloud native, and obviously I'm here to talk Kubernetes in containers. So the more I get to talk about that, the happier I'm going to be. Well, it's a hot space. We're bringing you the cube inside of our little container here. Dan, thanks very much for coming on today. Thank you. All right, Dave Vellante for Stu Miniman. You're watching theCUBE from IBM Think. Day one, we'll be right back right after this short break.