 From Seattle, Washington, it's theCUBE, covering KubeCon and CloudNativeCon North America 2018. Brought to you by Red Hat, the CloudNative Computing Foundation and its ecosystem partners. Everyone, welcome back to theCUBE's coverage here. Day three of wall-to-wall coverage at KubeCon, CloudNativeCon 2018, here in Seattle. theCUBE's been breaking it down all week. I'm John Furrier, Stu Miniman. Our next guest is Tuan Nguyen, who's the principal engineer in technical marketing, cloud products and solutions at Cisco Systems. Tuan, welcome to theCUBE, thanks for joining us. Thanks for having me, thank you. So obviously, Cloud has been a big part of Cisco. We've seen at Cisco Live last year in Cisco Barcelona, got a big European event coming up Cisco Live in Europe. Cloud has been big part of the CEO's conversations on stage. Cisco's going all in on Cloud. DevNet, DevNet create two communities. You guys got a cloud-native vibe going on at Cisco. Cloud Center, you got some products that are addressing this. This is a shift for Cisco, big time. You've been in the cloud, it's just like all, it feels like an all-in. Right, right, yeah, yeah. So what we've been evangelizing to people here is that Cisco is a software company, right? We certainly have a very strong heritage in our enterprise relationships related to our hardware platforms, but we're transitioning and we're really making that conversion to being a software company. Cisco has been acquiring talent and technology in the past couple of years. We've developed some strong relationships with Google and AWS as well, and we developed these reference architectures that our customers can buy as kind of a single unit and get the support that they need from us. Yeah, so. We covered your recent announcement with AWS. Really nice, elegantly designed Kubernetes strategy where using EKS over here, you got the Cisco stuff on here, so it's a seamless experience for the customer, which is great. Congrats, I think that's a great announcement. I think it's directionally correct. I think that's what customers want, but I want to ask you a bigger question. I want to get your opinion on perspective. When you look at Kubernetes, what we're hearing here at the show from end users and from the emerging startups that are contributing is that breaking down the monolithic application into a series of granular sets of services is what everyone's doing. That's clearly, it's microservices, a variety of other things. Kubernetes can connect that, but it's the network that brings it together. So we're seeing the policy knobs inside Kubernetes as being a very strategic benefit. We had one expert say, why don't you want to take advantage of those policy knobs? It's a great opportunity. You guys are as networked as you could be. It's Cisco, this is your DNA. How are you guys looking at Kubernetes? Are you looking at the policy knobs? How do you talk to your customers about this new opportunity with Kubernetes? What's the real upside for your customers with Kubernetes? Yeah, so one, you mentioned, we see Kubernetes as very pervasive, so we offer an on-prime version of Kubernetes, and of course, we partner with Google and with AWS to deliver on-cloud versions of Kubernetes. And related to policies, application policies in the form of Istio and network policies or security policies in the form of a network interface, our on-prem solution offers three types of CNIs. So we're very flexible in that way, and certainly if you are a Cisco customer and you have a Cisco ecosystem of hardware platforms, then we natively integrate into those platforms and we let you leverage your existing investments. So if I look at it that way, then I'm saying, okay, I'm good with Cisco right now. Do I have to change anything with Kubernetes? What's the impact of me as a Cisco customer? Is this added value, consistent environment? What's the impact of the customer's day-to-day operational environment? Yeah, so our customers are asking us to tie both VM-based and container-based workloads into CI CD. So we obviously, with our ACIC and I, we give them the capability to construct policies in Kubernetes that end up on the hardware platform, right, that's number one. And then we also have a harbor registry, we have security policies that can be carried across different platforms. So in your private cloud and VMware and OpenStack, you can carry those same policies. And for us, we've got application delivery frameworks and platforms that deliver the application. So in the form of both VM and container-based as well as bare metal. And we kind of unify the user experience when it comes to application deployment in Kubernetes. So, Twana, I'm actually, I'm glad that we got you towards the end of what we've been talking about here. One of the things we've been teasing apart is multi-clouds in many ways is like what we've been talking about a long time about multi-vendor. And the networking space is an area that we really understand what worked and what didn't work in a multi-vendor world. And the management piece was often the breaking point because just stitching all those together, we've looked for the last few years, customers have multi-cloud and getting their arms around that and how do I manage that, can be a real challenge. So, we know Cisco's making investments, they've made acquisitions. Tell us, what have we learned from the past? What's different about this now that will make it successful where management has been one of the pitfalls for quite a long time? Yeah, yeah. So I think what we've learned from the past is that customers are asking us for policies that can span across the multi-cloud, right? So, whereas certain platforms will give you a hybrid cloud experience, Cisco is investing in things like VPN mesh topologies in the CSR, in ASR, in protecting workloads as they move across different cloud targets. And then also in the provisioning and lifecycle management, we feel that customers want the capability to run applications in any cloud environment and under any type of overlay or underlay networking platform, yeah. Tuan, one of the things that, to talk about not only getting your arms around it, but there's multiple accesses that I need to optimize for. One of the ones, of course, sorting out is cost. So, where does Cisco sit in this environment? The big shift that I think was really highlighted for me last year, going to Cisco Live, is it used to be most of what I'm managing, I control. Today, most of the network and most of the environments that I'm in charge of, they're outside of my purview. We're doing that multi-cloud world. So, how do I make sure that I don't get myself in trouble with the CFO or have unexpected things come up? Right, right, yeah. So, I came through a software acquisition called Clicker Technologies and Clicker Technologies is that one tool that gives you that experience and allows you to see cloud costs. So, cloud costs from a hourly metered perspective, but also from a budgeting perspective. And we're adding additional components into our platform that gives you true costs for all of your compute, all of your network, your storage, your services like Lambda, and then also makes recommendations on the instant sizes that you need to use. We have policies like suspension policies that help our customers to save on their cloud bill. And in a lot of ways, the lifecycle management aspect of applications is something that differentiates us from other cloud management platforms. Talk about the cost side and the cost of ownership. I've always been talking about the cloud as the TCO or total cost of ownership changes a bit. What are some of the challenges that you've seen the customers are having that you guys are helping with? When you look at the integrating security, networking and application performance and management, because it's not siloed anymore, they're integrating together. That's right. This is a new dynamic. Right, right. What's the state of the art? What are you guys doing? Do you guys address that? What are some of the customer challenges and just what's your thoughts in that area? Yeah, so most of the time, there are two basic challenges to this. One is bringing the cloud economy into the private cloud consumption is something that our platform does. And then also being able to visualize all the costs, helping our customers to make good decisions about what types of workloads run where best and whether it's, so we enable obviously VMs as well as cloud native container-based microservices to coexist in a single platform. So we'll deploy VMs and containers in a hybrid fashion or we'll deploy them into the same and we'll give you the utilization of those workloads based on dollar amounts, based on runtime and also based on the type of workload. So here's the curveball question for you. Now multi-cloud comes into the equation. How do you guys deal with that? Because workload, so on behalf of some cases, I've heard from customers that refactoring those workloads is a problem. So if I'm going to run true multi-cloud and I have multiple clouds, I need networks to know, have smarts around where I want to put that and when I run it in different geography maybe or regions, so the network has the intelligence on a lot of things. How are you guys addressing the multi-cloud component with workload without refactoring? Yeah, so because we can compose applications that consist of both VMs and containers, right? One of the projects, one of the use cases that we worked on with our relationship with Google was from Cloud Center to deploy cloud native workloads in GKE that would navigate and basically traverse the VPN network to go back into the on-prem target in order to access a database that was kind of a legacy database using an API URL. So that whole workflow was something that we solved for with our reference architecture. So we obviously have the portfolio of products that allows our customers to take advantage of both hardware, software, and networking and security and monitoring all in one reference architecture. A lot of opportunities for you guys. I think you're positioned well. We've covered you guys on the DevNet, DevNet Create. You're seeing the Cloud Center, this dashboard kind of model of looking at the operation side, the development side, a lot of changes really kind of fit right into your wheelhouse. So I think the Kubernetes policy knobs, it's a big story that I'm walking away with on this trip is saying, wow, policy sounds like a networking thing. Networking guys love policy. You can automate it and manage the costs. It's a good thing. Thanks for coming on. Appreciate your insight. Thank you. Thank you very much. Cube Coverage here, day three continues. I'm John Furrier, Stu Miniman. Stay with us for wall-to-wall coverage here at KubeCon Cloud Native.com. We'll be right back with more after this short break.