 Live from Vienna, Austria. It's theCUBE, covering .NEXT Europe 2016. Brought to you by Nutanix. Here's your hoes, Stu Miniman. Welcome back to theCUBE's coverage of Nutanix .NEXT here in Vienna. It's getting towards the end of the day. Sunlight's starting to go down because it goes down to about 4.30 here in Vienna. I found here in the fall. Happy to welcome to the program a first-time guest. I was part of the keynote this morning. Tim Zonka is the Vice President of Product Marketing at Puppet. Tim, thanks so much for joining us. Thanks for having me here. It's great to be here. All right, so we've had, I believe it was, the founder and former CEO of Puppet been on the program, but tell us a little bit about your background. How long have you been at Puppet? So I've been at Puppet just over two years, so about two and a half years, and have a kind of history working either selling to or marketing to IT. And so it started kind of early days with security software doing sales engineering there and kind of ended up always gravitating toward kind of helping IT deliver better stuff to their users. So being at Puppet's been a blast because it's kind of our central focus and reason for existing, really. All right, so Tim, give us just kind of the real quick, you know, who is Puppet today? You know, certain people would go, oh, there's this Chef Puppet Ansible, kind of Ansible's bought by Red Hat recently. What's Puppet? Who differentiates it in its space? Sure, so in short, Puppet is an automation and orchestration platform that allows people to manage everything that they have in their data center in a standard way. And so we don't care if it's resources that you're running on the mainframe, if you have a development environment running Docker containers in AWS or you're running mission critical systems in something like Nutanix that we're here at the conference today, one language, one standard way for managing all of it. And you had to manage management change recently, Sanjay Mirchandani, KubeLum actually, back from his days at, it was EMC, even though it's the company that's now Dell EMC, still takes me a little while to wrap my head around that. I think you said he was there as COO and now he's CEO. Yep, he started in April, or I guess May as COO and president, and then recently Luke Kenees, our founder, handed him the reins as CEO. And so Sanjay's been a fantastic addition, especially with his experience in running one of the largest IT organizations on the planet when he was at EMC, and then also just with his operational background. So it's been great to have him join the company as we scale, especially around the world. And then he's brought on some other folks. We have Mike Roshan, who joined recently to head our people organization. And he had experience at Riverbed growing that. And then Gary Green, who's running our worldwide sales organization. Okay, so Tim, Nutanix talks a lot about simplifying IT. It's one click, everything. So maybe explain to our audience, why if Nutanix simplifies so much? Where does Puppet come in? What do you add to that equation? Sure, so I think that's part of a vision that we share, is being able to run your infrastructure, the middleware on top of it, the applications on top of it more simply. And it's not really just simple for the sake of simplicity, but ultimately in order to deliver better software faster to the people who are using the applications or the services that IT provides. I think one of the things that we're excited about working with Nutanix is combining some of the optimization and capabilities that they have, especially at the infrastructure layer alongside the orchestration and automation capabilities we have for doing pretty smart things with provision, all the way from provisioning through ongoing management of the infrastructure and we bring a lot of experience, I think, especially around the middleware and application stack for ultimately the solution that both of us would push out together then. Tim, I had one of Nutanix's customers on today and he said, I now have infrastructure that my applications love. You mentioned kind of the middleware and the application layer. How do you guys tie in between kind of that stuff that sits on top of the infrastructure and the infrastructure itself? Yeah, so one of the things we introduced about a year ago was the ability to define using Puppet's language, which traditionally has been used to define infrastructure, core services, think of things like NTP or DNS or operating system level configurations. What our customers have started doing with us is they've started moving up the stack and over the course of the last couple of years they've started managing things like their databases or their web servers, app servers and things like that. Many of them have started even defining full applications and then they use this for releasing and automating kind of the release of their applications. About a year ago though, we realized one of the things that we wanted to give them the ability to do was define these applications that span multiple nodes. So you may have an application that spans thousands of nodes being able to use the same language that they already had a bunch of these Lego blocks that they built for extending it so that they can now define services and applications that span thousands of nodes across a data center using the same sorts of tools that they had already gotten used to. And so we introduced that about a year ago and since then have been adding additional capabilities. And so now we have customers that use us from kind of core infrastructure up through middleware, up through full applications. These may be off the shelf, they may be ones that they've built themselves where they can use Puppet as kind of the common language and standard way for delivering all of it through that stack. Tim, a term you used in the keynote was portability. Could you kind of unpack for us what that really means? Yeah, I think one of the things that we see our customers doing is as they define the infrastructure, the middleware, the applications, whatever they're managing with Puppet, that set of services now becomes portable. Once you define what you want to manage in Puppet, it becomes really straightforward to say, well, I want to deploy this to a Nutanix environment. I want to deploy this to a public cloud infrastructure. I'm going to run this across a set of virtual machines in my private data center. And it becomes really easy to use that same set of code that you've described your set of services with and move it around. And so the result of this is we see a lot of our customers that end up using us to do either data center migrations, adopting cloud technologies, evolving their cloud technologies. And I think another thing too is, and I heard Sunil talk a little bit about this because they have the idea of kind of more, what I interpreted as mission critical, kind of stable applications, as well as those that have a bit more of this kind of spiking, it allows people who define their workloads in Puppet, they can move them across the different environments that would be used to manage those, those different sorts of workloads, either the mission critical production ones, or those that they may be either bursting to or just doing development and testing in. Yeah, one of the big challenge you have is, in the enterprise of course, is when you bring something new in, how do I manage that transition? Cause I'm probably not just, very few customers to talk to, just kind of get rid of everything. I bring something new in, I might migrate something over, I need to run both up and running. So it sounds like you guys help in some of making some of that easier. Yeah, we do. And I think one of the things is, especially as people talk about Docker or adopting Kubernetes, or especially not only experimenting, we're really seeing the first signs of moving those sorts of technologies into production workload, or workloads in production, running those sorts of new containerized environments. And to your point, almost every company you work with, they don't have the luxury of saying, oh, all this traditional stuff, let's just forget about it and just focus on the new world. They have to manage all of it. And so one of the things that Puppet allows them to do is again, manage it all kind of in a standard way. And so it allows them to kind of move to the future where they can adopt new technology at scale and in production. And then as it makes sense, and their workloads are defined in more traditional infrastructure, move those that make sense into some of the, the new world sets of technologies, like Nutanix, like AWS, like Docker, Kubernetes and things like that. Yeah, okay. So you brought up containers, actually something we haven't touched on much today, but what are you seeing from customers? What are some of the challenges just kind of container adoption? Our team was actually out at the Kubernetes conference in Seattle this week, at GoopCon. You guys had your show, we're talking about containers in the keynote here, Amazon, Microsoft, everybody's talking about containers. So where are customers with that? And what do you guys see as some of the market landscape? It's been a blast to be around because I think as the conversation around containers has shifted over the last couple of years, we've really seen the kind of the threat of production tarnishing the new shiny object that is containers. And that's actually a blast for us. That's where people come to Puppet to adopt a new technology. So if you take a look back, kind of in the rear view mirror of where Puppet's been used across our customer base, it's been to help them adopt new stuff in production. So five or so years ago, virtualization was the kind of the big thing. Then it was OpenStack. And now more than half the OpenStack deployments around the planet are run with Puppet. More recently, it's been Cloud and DevOps where people are either using Puppet to adopt Cloud technologies or at least evolve them. Containers I think is the most recent thing where most of our customers have moved beyond or they're starting to move beyond experimenting with containers and they're ready to start seeing, well, how does it make sense to use this new technology when I'm actually running my workloads kind of in production at scale? And that's where we come in. And so we've been working with a lot of our customers to figure out the right tools that you need to do that, the right orientation of those and the kind of the best practices for actually running containers at scale in production. You mentioned, you know, being in Kubicon, even a year or so ago, there was kind of DockerCon and a host of little meetups. Even the proliferation of the different technology-centric shows have popped up a ton over the course of the last year. And I think that's indicative of just this movement of, not only has that, I would say, the containerized environment really taken off, but in each of these discussions, you see the kind of the threat of production really coming into the conversation. It's been a blast to be in. So Tim, you guys had your conference, PuppetCon, recently. What were some of the top technology adoptions that people were looking at tackling? You named off a ton of things, public cloud, open stack, Docker. If you could stack rank for us a little bit, where are your customers right? I mean, it's two, it's cloud and containers. And so it was, and it was cloud in a, I kind of grouped cloud into a couple categories. There's more of the traditional set of cloud services and capabilities where it's really kind of, the platform of cloud. And then some of the newer conversations around cloud, around things like cloud functions, or like lambda and things like that. We're not seeing a ton of that quite yet. Most of the conversations tend to be around multi-cloud. I have five different clouds. What's the right way to manage all these? When do they make sense to do it in a consistent way? When does it make sense to kind of leave them in pockets? So that was one big conversation. I think the other was then around containers as well. And really what does it mean to move workloads to these new environments kind of in progressions? And it sounds even, those two things tie to each other. Because, you know. They do, yeah. I mean, in most cases people are spinning up containerized environments in the cloud. Not all, I mean, they're mutually exclusive, but yeah, they do go hand in hand. Yeah, so, you know, there's that, you guys handle some of the orchestration. I mean, management across multi-cloud is still, you know, pretty early. So, you know, do you guys have answers for the customers that are doing that? Because it is a multi-cloud world. Yeah, and I think it's a pretty straightforward and simple one, which is, and as I've mentioned before, you know, with Puppet as you define the infrastructure that you want to run, the middleware layers, the applications, we really don't care where you're deploying it. And so it allows people at least to get their first steps in kind of standardizing across a multi-cloud environment where they have a consistent set of workloads, a language to describe it all in a standard way for either provisioning or at least the ongoing management of it. And what we see happening now is we'll have a lot of groups or customers where they'll have groups that are especially using public cloud environments for a lot of their development or testing. And then ultimately moving them, at least today, a lot of the production workloads to some kind of private cloud infrastructure, that's even starting to shift. But what our customers find most useful is being able to kind of move these things around as they figure out the right places to optimize their usage and spend there. Tim, did you give session here in addition to the keynote? No, I haven't done a session. I'm curious if you've had the chance to talk to some of the Nutanix customers. I just, what's the overlap of kind of, you think a heavy infrastructure crowd versus where you guys sit in the stack? Yeah, it is a heavy infrastructure crowd. But where we also work pretty frequently with operations teams. And so these tend to be here at this conference a lot more on the kind of like the deep infrastructure part of those teams. And some of the conversations, one of the things that I've, I had a handful of conversations throughout the conference with is kind of the infrastructure and ops teams becoming more devy in the kind of a la dev ops world, which I, that was mentioned earlier this morning, but it's one of the things that people as they were coming by talking with us was, how does this, how do you help us get there? All right, so Tim, you set me up. I asked to ask the dev ops, ops dev, where those sits, that whole piece, what's your take on the discussion? You know, I actually, and I don't know if this is what you're asking, but like I don't think it matters. We have people ask us, you know, do you, is it start better on using puppet on the ops side or the devs, like it just doesn't matter. You got to get started. Our goal is, and I think working with our customers is to help them optimize the entire software production system. Both dev and ops are part of that. And so as long as you start somewhere and you're really optimizing that, if you start on the ops side, you're, you have to bring in dev at some point and vice versa, if you start on the dev side, you know, you can't optimize that entire production pipeline without including ops. Okay, Tim, want to give you the final word? Where do people usually start with, with puppy said everybody should start any kind of, you know, starting tips, you know, ways they can start down this path? Yeah, I think most of our customers find a lot of success in finding, so once you put puppet across your different systems, it shows you, well, here's all the things you have, so you get kind of situational awareness around what you have, and then you start to see the things that are changing. And so you could, most people start with really straightforward stuff, something like syncing your clocks. Doesn't sound particularly, you know, compelling, but it causes a ton of errors. ITP above, especially global IT, I mean Google worked on that for years. Right, and so usually they start with something like that or, you know, it's DNS or making sure their operating systems are consistent across the different systems that they're managing. Once they do that and have that kind of standard in place, then it becomes really easy for them to start to standardize then the middleware that's running on top of that and then start to take all these blocks that they've put together now and start composing them into applications. All right, well, Tim Zanka, it's a pleasure to talk with you. Thanks very much for sharing with our community. They're starting to roll out everything out for the happy hour, so we've got a little bit more here of our coverage of day one of Nutanix.NEXT. You're watching theCUBE.