 Live from London, England, it's theCUBE, covering .NEXT Conference Europe 2018. Brought to you by Nutanix. Welcome back to London, England. This is theCUBE's coverage of Nutanix, .NEXT 2018, 3,500 people gathered to listen to Sunil Poti. Give part of the keynote this morning. Sunil's the Chief Product and Development Officer with Nutanix. Glad we moved things around, Sunil, because we know events, lots of things move. Keynotes sometimes go long, but happy to have you back on the program. No, likewise, anytime. All right, so I've been to a few of these, and one of the things I hope you walk us through a little bit, so Nutanix's simplicity is always at its core. I have to say, it's taken me two or three times, hearing the new, the broad portfolio, the spectrum, and it's, okay, I've got the core, I've got Essentials, I've got Enterprise. I think it's starting to sink in for me, but it probably will take people a little bit time, so maybe let's start there. I mean, I think one of the biggest things that happened with Nutanix is that we went from a few products just 12 months ago to over 10 products within a span of a year. And both internally as well as externally, while the product values are obviously obvious, so it's more the consumption within our own sales teams, channel teams, as well as our customer base, needed to be codified into something that could be a journey or an adoption. So we took it customer-inwards, and about a journey that the customer goes through in adopting services in a world of multi-cloud. And before that, before you get to multi-cloud, you have to build a private cloud that is genuine, as we know. And before we do that, we have to re-platform your data center using HCI. So that's really, if you work backwards to that, you start with Core, which is your HCI platform for modernizing your data center, and then you expand to a cloud platform for every workload. And then you can be in a position to actually leverage your multi-cloud services. Yeah, and I like that. I mean, start with the customer first is where you have, and I mean, the challenge is, every customer's a little bit different. One of the biggest critiques of, you say, okay, what is a private cloud? Because there tend to be snowflakes. Everyone's a little bit different. We have a little bit of trouble understanding where it is, or did it melt all over the floor? So, it gives a little bit of insight into that, and help us through those stages of the journey, the crawl, walk, run. I think the biggest thing that one has to understand here is that these are not discrete moving parts. Core is obviously your starting point of leveraging computer storage in a software defined way. The way that Amazon launched with EC2 and S3, right? But then every service that you consume on top of public cloud still leverages computer storage. So in that sense, Essentials is a bunch of additional services, such as cell service, files, and so forth, but you still need the core to build on Essentials, to build a private cloud. And then from there onwards, you can choose other services, but you're still leveraging the core constructs. So in that sense, I think both architecturally, as well as from a product perspective, as well as architecturally from a packaging perspective, that's why they're synergistic in the way that things have rolled out. Okay. So looking at that portfolio, so a lot of customers I work with, the customer I work with now, they don't start out in the data center. They've already moved past that, right? So they are leveraging a partner, the public cloud. They might not even be running virtual machines at all anymore. How does that fit into your portfolio? Yeah. I mean, increasingly, what we're realizing, and we've done this over the last couple of years, is for example, with Calm, you can only use Calm to manage your multi, sort of public cloud without even managing your private cloud of Nutanix. Increasingly with every new service that we're building out, we are doing it so that people don't have to pay the strategy tax of the stack. It needs to be done by a desire of I want to do it versus I need to do it. So with Frame, you can get going on AWS in any region in an instant, or Azure. You don't need to use any Nutanix on-premise. Same thing with Epoch, with Beam. So I think as a company, what we're essentially all about is about saying that, let us give you a cloud service-like experience, maybe workload-centric, if it is desktops and so forth. Or if you're going to be at some point reaching a stage where you have to replatform your data center to look like public cloud, then we have the core, if I can call it platform itself, that'll help you get there as well. So looking at replatforming that data center, if I were to do that now for a customer, I wouldn't be looking at virtual machines, storage, networking. I'd be looking at containers, or serverless, or the new stuff. Again, what is Nutanix's answer to that? Yeah, I think what we've found is that there's quite a bit of an option, obviously of cloud-native apps, but when it comes to mainstream budget allocation and all, it's still a relative silo in terms of mainstream enterprise consumption. So what we're finding out is that if you can leverage your well-known cloud platform to not create another silo for Kubernetes, don't create another silo for Edge, or whatever the new use cases are, but treat them as an extension of your core platform. At least from a manageability perspective and operational perspective, then the chances of you adopting, or your enterprise adopting, these new technologies becomes higher. So for example, in Calm, we have the pseudonym called Calm with a K, right? Which essentially allows Kubernetes containers to run natively inside a Calm blueprint, but co-exist with your databases inside a VM because that's how we see the next generation enterprise apps morphing, right? Nobody's going to rewrite my whole app. They're going to maybe start with the web tier and the app tier as containers, but my database tier, my message queue tier is going to be as VMs. So how does Calm help you abstract the combination of containers and VMs into a common blueprint, is what we believe is the first step towards what we call a hybrid app. And when you get to hybrid apps, is when you can actually then get to eventually over a period of time to cloud native apps. One of the questions I was hearing from customers is they were looking for some clarity as to the hybrid environments. The last couple of shows, there was a big presence of Google at the show, and while I didn't see Google here on the show floor, I know there was an update from kind of G8, GCP, and HV. Is Google less strategic now, or is it just taking a while to, you know, incubate, you know, hunt and death fit? The way that you will see us evolve as we navigate the cloud partnerships is to actually find the sweet spot of product market fit with respect to where the product is ready and where the market really wants that. And some of it is going to be us doing, you know, a partnership by intent first, and then as we execute, we try to land it with honest product. So where we started off with Google, as you guys know, is to actually leverage our cloud platform, Zai, co-located with Google data centers. And then what we've evolved to is the fact that our data centers can, quote, unquote, integrate with their data centers to have a common management interface, common security interface and all. But we can still run as co-located ones. Where the real integration that has taken some time for us to get to is the fact that, look, in addition to COM, in addition to GKE kind of things, is rather than us run as a power sucking alien on top of some Google hardware, the true integration comes with us actually innovating on a stack that learns AHV natively inside GCP. And that's where nested virtualization comes in. And we've had to take the crawl-walk-run approach there, which is because we didn't want to expose it to public customers what we didn't consume internally. So what we have with the new offering that we've announced called Test Drive is essentially that. We've proven that AHV can run in a nested virtualization mode on GCP natively. It can co-locate with the rest of GCP services. And we use it currently in our R&D environment for running thousands of nodes for pretty much everyday testing on a daily basis. And so once we've exposed that now as an environment for our end customers to actually test drive Nutanix as a fully compatible stack go on premise. You have Prism Central, the full CDP stack and so forth. Then as that gets hardened over a period of time, we expose that into production and so forth. So there's one category of cloud I haven't heard yet, and that's the service providers. So Nutanix used to be a really good partner for service providers for enabling them to deliver services locally to local geography, stuff like that. So what's the sense of Nutanix regarding the service providers? Yeah, I think, frankly, that's probably a 2019 material change to our roadmap. It's your, the analogy that I have is when we launched our operating system, we first had to do it with an opinion stack using Supermicro. Most importantly, from an end customer perspective, they got a single throw to Joe. But also equally importantly, it kept the engineering team honest because we knew what it means to do one-take upgrades for the full stack. Similarly, when we launched XI, we needed to make sure that we knew what SREs do at scale. And so that's why we started with our version of SMC on, as you guys know, with digital reality as well as partners like Sixterra. But very soon, you're going to see us that once we have created that opinion stack, software-wise, we're able to leverage it. Just like we went from Supermicro to Dell and Lenovo and seven-hour partners, you're going to see us create a XI partner network, which essentially allows us to federate XI as an OS into the service providers. And that's more a 2019 plus country. Yeah, speaking along those lines, in the keynote this morning, Carbon with a K, talked about Kubernetes. Talk about that, that's the substrate for Nutanix's push towards cloud native, so. Yeah, yeah. I mean, I think you're going to hear that in the keynote, they took keynote as well, as basically, you know, customers want, as I said, an operating system for containers that is based on well-known APIs like KubeCuttle from Kubernetes and all that. But at the same time, it is curated to support all the enterprise services, such as volumes or storage, security policies from Flow. And, you know, the operational policies of containers shouldn't be any different from VMs. So think about it as the developers still get at Kubernetes-like interface. They can still port their containers from Nutanix to any other environment. But from an IT ops side, it looks like Kubernetes, you know, containers and VMs are co-residing as the first class option. Yeah, I felt like there'd been a misperception about what Kubernetes is and how it fits. You know, my take has been, it's part of the platform, so there's not going to be a battle for a distribution of Kubernetes because I'm going to choose a platform and it should have Kubernetes and it should be compatible with other Kubernetes out there. Yeah, I mean, it's going to be like a feature of Linux. Yeah. In that sense, there's lots of Linux distros, but the core capabilities of Linux are the same, right? So in that sense, Kubernetes is going to become a feature of Linux or the cloud operating system so that those least common denominator features are going to be there in every cloud OS. Okay, so Kubernetes not differentiating, it just expands the platform. Enabling. Yeah, yeah. Enabling piece. Tell us, what is differentiating today? You know, what are the areas where, you know, Nutanix, you know, standalone is different from some of the other platform providers today? Yeah, I mean, I think the, I mean, obviously, whatever we do, we are trying to do it thoughtfully from the operational, you know, simplicity as a first class citizen, like how many new screens do we add when we need new features? A simple example of that is when we did micro segmentation, the part was to make sure that you can go from choosing 10 VMs to grouping them and putting a policy as soon as possible with as little friction of adopting a new product. So we didn't have to virtualize the network. You didn't have, you didn't need to have VX LANs to actually micro segment, just like in public cloud, right? So I think we're doing the same thing into services up the stack. A good one to talk about is ERA, which is essentially looking at databases as the next complex beast of where, you know, operational complexity resides, especially Oracle RAC. And it's easier to manage Postgres and so forth, but what if you could simplify not just the open source management, but also the database admin? So I would say that ERA would be a good example of a strategic value proposition of what does it mean to create a one plus one equals three value proposition to database administrators. Just like we did that for VMware administrators, we're now going after DBAs. All right, well, Sunil, thank you so much. Wish we had another hour to go through on it, but give you the final word as people leave London this year, you know, what should they be taking away when they think about Nutanix? Yeah, I mean, I think, you know, the platform continues to evolve, but the key takeaway is it's a platform company, not a product company. And with that comes the burden, as well as the promise of being an iconic company for the next hopefully decade or so, all right? Thanks a lot. Well, it's been a pleasure to, you know, watch the continue progress. Always a pleasure to chat, Sunil. Thanks a lot. All right, for you, Piskar, I'm Stu Miniman, back with more coverage here from Nutanix.NEXT 2018 in London, England. Thanks for watching theCUBE.