 Live from Las Vegas, it's theCUBE, covering AWS re-invent 2017, presented by AWS, Intel, and our ecosystem of partners. Welcome back to the stands. We're here in Las Vegas, just off the strip as the re-invent show continues here with a very exciting day one. You talk about buzz on the show floor. The place has been jam-packed since they opened the doors at 11 o'clock, our time this morning and continues to be so, and I imagine for the next two or three days you're going to see a lot of people, 50,000 plus, a lot of exhibitors, a lot of people, a lot of buzz, a lot of excitement here around the AWS community. We have with us now Chris Wall, who's the Chief Technologist at Rubrik, and he knows so much about this space that takes three hosts to surround him. John Wall's here, Lisa Martin to my left, Justin Warren on the far right, you're surrounded. I am, you've ganged up on me. Yeah, right, well, and rightly so. Yeah. So you're, Justin can't wait to hear. He brings the evil eye from Justin. This feels like a trap. There's some good history here going on, so we'll find out a little bit later on. First off, Chris, do welcome, welcome to theCUBE. Tell us a little bit about Rubrik and what you guys, your place here and your feelings about the show. Yeah, yeah, so Rubrik, about two and a half years in the market, about three and a half years old as a company, really focused on solving the conundrum around, there's all this public cloud stuff out there and everyone's kind of feeling the elephant with the blindfold on, describing it differently. And we're trying to figure out how can we take that cloud type architecture that's out there in the world and combine that with a almost $50 billion TAM that is data protection, backup recovery archive, put those two together to solve challenges they've enterprise is really struggling with, onboarding into the cloud and using those resources as well as making sure their assets be at the application, the data itself or a physical server be protected and available for recovery in a really, really quick way. So that's kind of the high level pitch of Rubrik around the last 10 major releases of the product and it's been a rocket ship, I've really enjoyed it. So bigger focus in enterprise or are you also playing with the startups and helping them also transition? That's a good question. I mean, originally we're kind of looking mid-market, you know, like, okay, let's go for that kind of sweet spot but very early on, a lot of large enterprise customers came up and said, wow, you're fully arrestful API compliant, the full stack is distributed and scale out and really solves their problems. So they kind of pulled us into that space and ever since we've really embraced the large enterprise globally. It doesn't really matter where you are in the world. Those are challenges that are kind of ubiquitous across verticals and the market. So I've got a good storage and backup background as you well know. Okay. There has been such a big shift in data storage and backup and data protection in the last, say, three or five years. What do you think is driving that? Because really, like backup and recovery was always a hygiene function. It was boring, no one really wanted to spend any money on it but now you're part of this new guard of brand new ways of doing things that has that part of the market being kind of exciting again. I almost feel like we got used to the horrible nature of that business because as a technologist, I was a customer for about 13 years, I was in the channel for about five and it was always, well, this is just the way it is and you got to put up with slow restores that are clunky, it was seen as an insurance policy. I think as the enterprise mature to the point where everything else was amazing and hyper-converged and driven by APIs and cloud is starting to eat up part of the data center, we finally saw, okay, this isn't going to stand. We can't operate in a model where an RTO is days or even many hours and it's really heavy lift. I needed a full team of people to manage this stuff. So I think as the technology advanced as well as kind of outpaced all the data protection software and solutions are out there, it just kind of had to happen. And I think as cloud and object store also permeated in the market, it really gave us a great opportunity to use that for long-term retention beyond just the old tape and things like that. Yeah, okay. Yeah, that sounds fair. So what do you think has been the biggest cultural change? Because I mean, this is lots of technology that goes into that, but you're talking about having whole teams of people who have to herd this stuff around with small little toothbrushes and stuff just to keep the thing running, whereas now you can pretty much run it with one or two guys just sitting there and just go, yeah, it just works. Well, it's similar to, remember when we went through virtualization and it used to be whole armies of people managing all these pizza boxes and two-yube servers and there was just a lot of infrastructure and operations people necessary to run the data center and then we virtualize and I know my personal story was there was two of us managing 1300 virtual machines. Right, so that scale is astronomical compared to what we're used to. Well then apply that kind of mentality to data protection and it's yeah, it's a few minutes from one person or a distributed team that spends a few minutes a week, maybe a month, something like that, managing things more at the policy and the tag and the metadata layer and it's that journey all over again. So the nice part is we've done this before, we know it can be done, but kind of the hard part is people are always the hardest part of the equation and it's sometimes the stuff to put your hands off the handlebars of the bike and just say, I trust an intelligent system to manage this part of the stack and I'm going to go focus on where are we trying to go? Speaking of trust, you talked about how your enterprise customers have pulled you into or up the chain there. A lot of what Andy Jassy said recently to John Furrier is $18 billion run rate growing at 42% a year. They haven't gotten this big with just startup salons. So he's talking about enterprises really being on the precipice, this mass migration. How does being a young company, how does your relationship with AWS help give more credibility to Rubrik as a trusted advisor to these enterprises? Yeah, I'll kind of start at the end and then work my way backwards. So we recently hit the advanced tier partner status with Amazon and part of that I would cite a couple of public references with Castilea schools as well as Fuji radio are two different companies in different parts of the market but they're very much focused on we need a partner that can bring us into the cloud kind of onboard us into that environment. AWS was the specific cloud provider they were looking to get into without kind of operationally, that's a scary thing. You know, it's tough as an infrastructure or operations focused engineer or even as a developer I think sometimes to say I want to take this data and it represents my apps and my servers and my solution stack and put that into public cloud even either for archive and retention or potentially to use our cloud instantiation solution that was recently announced where they can start building workloads into public cloud. And so I think that's why kind of at that point we work backwards a little bit to say as we work with the customers that were looking to do that before it was, well, you have to learn all this stuff and really become super technically deep on it. And I love that article by the way I thought it was really deep but if you look at this week in AWS that by Quinnipig on Twitter he's always pointing out every week the S3 bucket failure because it's hard cloud is really, really hard. So if you have a kind of that abstraction layer that can make it really simple for customers to onboard in there and make it, it's like simple but it's also abstracted from the nuances across multiple public cloud providers including AWS. I think that's the magic sauce that really gets people excited about it. That abstraction also probably gives them a little bit more comfort, right? Exactly. Some of the sausage making they don't have to see. Exactly, because part of our secret sauce is as the data is entering into that environment we're not just saying okay it's there, done it's now your problem. Part of our cloud data management story is that the data enters that environment but we're constantly checking it making sure it's valid, making sure that it's secure. We handle all the encryption, the data efficiency. You know the whole end to end life cycle of the data is respected whereas traditionally it was you just kind of scrape data at the data center you drop it off into an S3 bucket you pray that it's going to be there when you need it and who knows? Now it's IT-opsis problem. We don't just do the handoff and say good luck. We handle it from cradle to grave for all the data. You mentioned a little bit ago you're talking to Justin you're talking about the horrible nature of things four or five years ago, right? So no matter what time you're in there's always a horrible nature of things. There's always a problem. So now that whatever was the issue then what is the issue now? I mean as you new capabilities are developed it opens a whole new Pandora's box of challenges and problems. You have unforeseen issues. So what do you guys, when you're looking at your headlights, 12, 18 months down the road you say oh yeah, this is our next one we got to tackle. This is the baby we got to get our arms around. I mean for me more near term it's around the transition from trusting infrastructure to provide high availability and disaster recovery and moving that more towards the application and the stack itself. So holistically in the past you'd have two data centers they replicate, one's for DR, one's not. The cloud wasn't really in that equation and all of the redundancies was handled at the infrastructure layer. Well okay now if we can kind of surround metadata around the application, provide instant search, global availability, replication, the ability to actually stand up those applications in a public cloud. Well now the question is do I really need that infrastructure layer anymore? Do I need the second data center? Can't I just use public cloud or an MSP or someone that's providing rubric as a service as an example to provide that for me? More long term I tend to look at kind of in the discussion that I saw between John and Andy Jassy was around the part where I get really nerdy is around serverless and the ability to provide functions kind of in the data path. And now I start to imagine okay we're putting a lot of data for customers into public cloud and even kind of private object store resources and there's the ability, I think it was Greengrass is an example where you can kind of put that shim layer into the edge to do the function as the data's going in there. There's a lot of interesting opportunities that I'm looking forward to in the next year where well we already have an index of the data, we already are very cognizant and content aware when it comes to what we're protecting. Wouldn't it be cool if we could do more interesting things with the data in flight as well as where it's ultimately arresting, kind of like with the announcements with the media and the transcoding and the video services that I think came out rather recently. So that's kind of the two stage answer to that question that I have. Yeah so Chris one of the ways that AWS has succeeded is by appealing to developers. And you're talking there about things that are in the application layer that have nothing to do with infrastructure and developers hate infrastructure. So what are some of the things that you're doing that Rubrik is doing to appeal to developers specifically in being able to access their data and not have to manage it, as you say, the way we used to do it, which was the very infrastructure centric problem. How are they, what are you using to expose the data and to manage it as a data problem rather than an infrastructure problem? Well I think that goes back to traditionally how we managed infrastructure, especially on-prem. And it was all very manual, very imperative, meaning you're pulling the lever and you're telling the system, it's a dumb system that you're the intelligent layer of it and you have to control it. It doesn't work in the cloud model at all. And it really, I don't think works long-term in the data center model because then I always have to scale literally people to data. And that doesn't work. Humans don't scale. We can't just get magically more of us. So what we've done differently from day one was design a system where every component within the stack, even internal communications are calling RESTful APIs and the whole system is distributed. So there's no controller that you have to deal with. You don't have to become, you don't have to know anything about storage to use the product. It's not infrastructure bound. So you're able to control it completely through RESTful APIs or through configuration management tools, cloud management platforms, et cetera. So if you're a developer looking to, all right, I have an application. I want to make sure that it's automatically protected as part of that process and sent to AWS and automatically build me a cloud instantiation in EC2 as an instance. Great, make one, two, maybe three API calls. You don't have to know anything about infrastructure, which is the panacea. No developer wants to like dig into VLANs and things like that. That's really cool. And it solves a very valid business case in that if one person can write the code and it works, just repeat that process and it scales infinitely. I don't need extra developers for that. So to be able to do that I need to understand that API exists. So what are you doing to actually show developers that hey, this thing is here and this is how it works. He's something that you actually know how to do. How are you exposing that idea to the developers? Very early on we worked with Swagger which ultimately has become the open API spec 2.0. And so every node within our distributed system actually surfaces the entire API suite in two formats. One is like a playground. So you can, even if you're newer to APIs maybe on the infrastructure side you can kind of do a try it now button. You can kind of say what would happen and it surfaces what the call looked like and how to structure properly and what the return codes are. But more importantly, there's also the why and the how of the API and in a different kind of documentation suite using Redoc where you can go in and literally see okay, what's the mindset here? What's the use case? What's the example? And I feel like that's typically what's missing in a lot of these equations where it's just here's the nuts and bolts, here's the tactical information, here's push this button things happen. It's more like here's why you would use it, here's an example and a lot of the code to do that is already been created by our ranger team internally and made either exposed publicly as open source or privately as something that we share with our customer base. Cool? Yeah. Well, Chris you described you said a rocket ship, right, you've been on for two and a half years. I think you better fasten that seat belt. It's not going to slow down for you, I don't think. I appreciate that. Which is a good thing, right? Yeah, yeah. Yeah, it's all good. Yeah, I know. Hope you didn't feel ganged up on either, right? You came out of this, okay. I'm pretty proud. I appreciate that. I think so. Chris Wall from Rubrik, joining us here as we continue our coverage live here on theCUBE where AWS is reinvents live in Las Vegas back with more in just a bit.