 from Las Vegas. It's theCUBE, covering AWS re-invent 2018. Brought to you by Amazon Web Services, Intel. And their ecosystem partners. Welcome back to the Sands Expo. We continue our coverage here on theCUBE of AWS. Re-invent our seventh year at the show, getting bigger and better than ever. John Walls, Justin Warren joined now by Anthony Lai who's SVP and GM of Clouds Data Service at NetApp. Anthony, good to see you, sir. Thanks for joining us. Thanks very much. Very good, and Matt Baldwin, who is the director of Cloud Native Engineering at NetApp as well. Matt, good to see you this afternoon. Nice to meet you. Lunch good for you guys, by the way. We're just getting back into the swing of things here. Hope everything was fine for you on the show floor. Perfect. It's been a great, great show so far. Yeah, give me your take on what you've seen so far and kind of what you see, what you guys see as maybe the hot topic that you keep hearing a lot about so far this week. No, I think at AWS you're always sort of interested in what Andy brings onto the stage. He can fill three hours pretty comfortably with a lot of innovation. Just further reinforces I think the very fact that most of companies sort of, their intentions to digitally transform are happening in the cloud. I think Andy continues to sort of further support hybrid cloud as a reality. And so those architectures for us present enormous opportunity. So, tons of innovation, some on Kubernetes. I'd like to see a little bit more from AWS. But hybrid cloud I think has been the highlight for me. Okay, Matt, your take? I'm pretty much always focused on Kubernetes myself. So, the T-shirt is a bit of a giveaway. This is actually a collaborator shirt. So if you commit it to Kubernetes, you get the shirt. The container marketplace, I think it was the most interesting one for what I focus on. And I was really excited that we were part of that announcement. That we're part of the Amazon Marketplace as of this week for our NetApp Kubernetes service. So that's why, that's my big takeaway is that the container marketplace, that's why I'm, and the robotic service looked really damn cool too. That did look very cool. But tell us a bit more about Kubernetes and its use in the way that you're using it with NetApp. I mean, we were talking just before we went on air that people who are trying to write applications that need to manage state, that's always been quite tricky with Kubernetes. But you can now actually do that with, I believe it's the cloud volume service. Yes, so we just finished integration of integrating the old StackPoint technology that was acquired into the NetApp technology. So now we actually allow you to stand up a Kubernetes cluster, add Trident, which also exposes Amazon Cloud Volume Service, and then you're able to actually consume those disks as if you were consuming any other type of disk for all of your stateful workload. Because there is that meme in Kubernetes where you hear a lot of it, stateless only. Stateful, it's bad for stateful. And with NetApp combined with Kubernetes, we've actually solved stateful workload problems. So your data is production ready with NetApp's Kubernetes. You know, and what's nice is, you know, cloud volume service, unlike any of the other storage services you can get on the clouds, is truly delivered as a service. So you just provision an amount of a file system and you provision throughput and we NetApp run it for you. There's no, you know, AMI to install. There's no, you know, LUN to configure. It really is like dial tone. So when you combine a sort of a high performing storage service with a container based service, it enables the developers to do what they should be doing, which is developing and freeze them up from the things they shouldn't be doing, which is managing, monitoring, maintaining, upgrading. So NetApp sort of provides these services to its customers very much in the same way that, you know, Verizon provides dial tone. You don't understand how it gets there. You just expect it to work and that's what we do. Yeah, it should be about, developers should forget about how they manage the operational environment. The operational environment should at the end of the day be managing itself. And then all you're worrying about as a developer is how do I ship code into this environment? How easily do I manage that application once it's inside of the environment? So that's the other thing that we're doing with NetApp Kubernetes services, we're allowing you to not only do life cycle management of the infrastructure, life cycle management of the apps themselves that you've deployed into the environment. Okay. Now we feel that's very important. Yeah, auto scaling, self-healing. Yeah. All those kinds of services that application developers today have to figure out for themselves. Take a snapshot of your app if you wanted to. Yeah. Clone it, things like that. Back it up. Yeah. So for customers who are looking at transforming the way that they're doing stuff and embracing hybrid cloud or just moving to cloud, maybe they're already existing NetApp customers who haven't tried out some of this stuff. What does the journey look like for them? How do you help them make that transition into using that more of the just, just develop it? Don't worry about too much. Yeah. That's sort of four or four use cases that we see. There's sort of use case one is I have IT infrastructure and I want to extend it to public cloud. So I'm going to deploy disaster recovery. I'm going to deploy backup. I may deploy some simple web-based applications onto cloud. And we have a solution called cloud volumes on tap for that that allows you to run on tap in the cloud just as you run it on premise and build a data relationship. For people who then say I've got to move the big primary, that's where we position cloud volume service, SAP workloads, Oracle database workloads. Then the third use case is this refactoring build new which we believe increasingly is going to be based on Kubernetes. And so combining our high performance storage services with Kubernetes for orchestration of infrastructure and application was a natural for us. The fourth scenario we see of course is what we call cloud analytics, which is how effectively am I running my infrastructure on cloud, hybrid cloud? Am I cost models right? Am I placing my workloads in the right location? And we introduced about a month ago a new product called Cloud Insights, which is that sort of cloud analytics performance monitoring and charge back platform. Yeah, it sounds like you're answering a lot of key questions, right? So customers were coming to you and looking for those kinds of insights, I assume that's what drove that. You know, it's amazing to think really that, you know, NetApp decided to embrace the cloud about four, four and a half years ago and ported its flagship operating system from its engineered systems into AWS. And I think most people at the time predicted the end of NetApp and the end of on-premise and anybody doing anything in storage would be dead. And here we are, you know, with a huge booth showing 100% software, 100% running inside AWS. We're one of AWS's biggest partners. Thousands of customers rely on NetApp, whether they're an existing NetApp customer on-prem or not. Yeah, I think a lot of the community didn't count on is that data was the big problem for Kubernetes. It always had been. Always from the day one, it was data. And NetApp was able to actually solve that data problem. Yeah, it's funny how people forget about the information that lives inside these applications. People actually care about these sorts of things. Well, you know, you started for everybody, everybody likes to talk about the really sexy things. Cassandra, cool. What about Postgres? Most apps use Postgres or MySQL. I love Postgres. Postgres is great. Yeah, I mean, you know, like when we look at metrics, we see most of the usage is Kafka, Postgres, and MySQL. Yeah, yeah, data's heavy, right? Yeah, people are talking about this movement and all this infrastructure spinning up over here, through over there, and I'm like, where's the data? And they're like, oh no, how do I get those terabytes from here to here? And that's a question. So you have a lot of insight into the way that customers use data because you've been looking after their data for so long. So what are customers doing? Are they just spinning up new workloads in a new environment and leaving the data there? Are they actually moving data from one place to another, or do they pretty much build their applications around where the data already lives? It's really a combination. I would say, you know, there are some workloads where the data is largely controlled exclusively by the application logic and it sits there. Increasingly though, people realize that data by itself is of some value, but when they move the data and merge it with some other data, there is sort of an enormous amount of derivative value that they can get from that. So you heard Andy talk this morning about data lakes. Everybody's jumping in to data lakes. I think that's very true, and therein lies an enormous problem that your data has largely been siloed. It was hard enough to manage your data when it was in your physical data center and as great as Amazon does, it just created another place to move your data to. So now your data's not within a few meters of each other. It's now potentially hundreds of miles away and the internet is only so big and data is not very nice to move. It doesn't like to be moved. So a lot of people are moving data backwards and forwards where it makes sense. A lot of people struggle with it and I think NetApp provides solutions to support both in place data models as well as data orchestration and data integration. So did we give people capability that they weren't sure how to use? Is that, I mean, I hate to... No, I wouldn't say. I think there's obviously a learning and I think a lot of people come here to get an education. All right. I don't think everybody here wouldn't be live on AWS. There's a lot of people here that are saying, what does it mean for me? What kinds of things should I do? And so in some cases, yes, there is that sort of, we have to explain strategies to some of our customers. What we want to do at NetApp is try and provide a fabric, a sort of a utility that allows freedom of choice. At the end of the day, NetApp will survive if we allow our customers to manage their data as effectively as they want. And we believe a big part of that is liberating it. We don't want people to consider NetApp a closed environment, which is why we embraced AWS. We love it if our customers see NetApp as a facilitator to use all of these wonderful services. Matt? Yeah, I'm starting to see more and more users begin to span workload. So you got two types of workload. You got Greenfield, Brownfield. At the beginning of the Kubernetes space, you started to see most all Greenfield apps moving into Kubernetes. So anything that you're writing that's new. So that's where data, you get to choose where your data's going to live now. But most of the data is Brownfield. Most apps are Brownfield. So how do we extend into that? That's why we're doing work around things like Istio to begin to expand your service mesh so you can incorporate in Brownfield applications that now can participate inside of a Kubernetes cluster so that you can actually stand up services and migrate services that are not attached to that data into K8s and then start to solve the problem around data. Maybe you want to do data tiering up into Amazon from your on-prem. You're able to do that. So that's where I'm starting to see things. I mean, I think, you know, Kubernetes itself was, I think, a fantastic step forward. Istio, I believe, is of equal if not more significance to the world of cloud, hybrid cloud, and the whole construct of these new applications that can consume these services and ultimately data. I mean, you can have, the beauty of Istio as a service mesh is that I'm able to actually have a service that's living in site A that does need access data. That data could be in site B and I'm just routing the traffic to go get that data. So I'm able to do things like that with the product and that's where the power of all this is coming in. That's why, as you go from monolith to microservice pattern, that's, you know, now you're able to start moving pieces out into other environments. Well, judging by the booth, and as you said, the presence is strong, vibrant, and you are alive and well. Right? We're having a bowl. I thought it's great to be a NetApp. We appreciate the time, gentlemen. My pleasure, thank you guys. Thank you for sharing that with us and your insights as well and good luck down the road. Thank you. All right, Anthony, Matt, thank you for being with us. Back with more from AWS ReInvent. We are live, we're on theCUBE here in Las Vegas.