>> Announcer: Live from Austin, Texas. It's The Cube. Covering DockerCon 2017, brought to you by Docker and support from it's ecosystem partners. (techno music) Hi, I'm Stu Miniman and this is Silicon Angle Media's production of The Cube, the worldwide leader in enterprise tech coverage and this is DockerCon 2017. We're here at the Austin Convention Center, just had the day two kick-off at the keynote. Really, yesterday was the developer day, today is the enterprise day. And to help me break down the latest news and what's happening in the ecosystem, I grabbed just some guy. (laughter) And of course, that's actually in his Twitter bio, which is why I do this, and I happen to have a good friend of mine, a good friend of the community, Stephen Foskett, who is the organizer of TechFieldDay. Stephen, always great to see ya and thanks for taking time out to get a little casual and dig into some open-source developer stuff. >> Yeah, you know, these are the developers, I'm used to wearing my fancy clothes, but I figured I would try to blend in a little bit here with the DevOps crowd at DockerCon. >> Yeah, I saw one of the demo guys had like a flashy jacket. I figured you'd come in in tails and- >> Yeah, I do usually have flashy shirts and stuff on, but yesterday I felt a little out of place, I mean these guys are, well, a lot of t-shirts here. >> Yeah, so today not as many announcements but it's always interesting. Shows like Amazon, shows like this, it's like, okay, one day let's talk to the developers and one day let's talk to the enterprise. What's your take on that? How is Docker doing with their maturation and what do you see in the marketplace? >> Yeah, I think that's really the key to what they're planning, so yesterday, I don't want to say developer because it was developer and ops but it was basically traditional Docker day, yesterday. And today is all about the enterprise. And I think that Docker had a very clear goal from today and that was to really plant their flag and say, not just Docker Day the center like last year, but that Docker is not only ready to be in the enterprise and not only has the tools to be in the enterprise, but is already there with some major customers. >> Yeah and great customers, had Visa and MetLife up onstage and no better way to say we're ready for enterprise applications than say, hey, Oracle is in the store there. What's your take, anything on the customer case studies Oracle? >> Well, let's take the customer case studies first. So clearly the takeaway from the Visa presentation and the MetLife presentation was nothing more than Visa is using Docker and Anchore, MetLife is using Docker and Anchore. I mean, basically these are massive, traditional companies with absolutely critical workloads, huge security requirements, and they're using Docker in production. I think that, if we would have all listened, Ben could have stood up there and said, "Hey everybody, Docker, MetLife, enterprise, production" and that would have been a substitute for 45 minutes of discussion. Because it's not like Visa's really going to tell us the secret ins and outs of their infrastructure, but they told us the most important thing, which is that a lot of those transactions are running through Docker containers. And that's what Docker wanted us to hear. >> It's interesting. Ben kind of blew up the myth of bimodal IT. And one of the things we'd kind of been looking at and want to get your opinion on, is taking my older applications and just kind of wrapping and moving them. Without changing a line of code, I can bring this into this environment what, you know many of us called for years, 'lift and shift'. What do you think about the modern, building new applications versus the old applications and, of course, customers don't have two IT environments. They usually need to move things together and have kind of a whole strategy. >> Yeah and well, I'm ambivalent about this whole concept of bimodal IT, but I'm not ready to reject it. I think it still matters from an app perspective, from an app-to-app perspective, and I think it's absolutely true that there are multiple kinds of apps. In fact, I think there's probably more than two kinds. I think that's maybe the real problem. You've got the real traditional applications, you know Southwest just announced that they're moving their reservation system forward from some old mainframe to some new mainframe, and that's causing all sorts of disruption in travel. Those kind of applications and then there's the more open systems packaged applications from the '90s and the 2000s and those things can be moved forward. And then there's sort of the applications that can be really modernized with containers and then there's the applications that you can 'microservice-cize' and then there's real cloud applications. So it's not just bimodal IT, it's really octomodal IT. >> And I like that Ben put it up there, it was a journey that they talked about. It's let's get everything on kind of a shared platform and have a way that we can do it the old way, start breaking it apart into more pieces, or totally rewrite. Because we know the migration cost of having to rewrite an application, it's really tough. >> Stephen: It's huge. >> But it's something that, for too long, people were like, 'oh well, I'll just run on that really old application that kind of sucked for way too long,' so I know sometimes I'm getting on my soap box and being like, please, your users hate that application and they'd like to be a little bit more modern. But it's not an easy thing and there's multiple paths to get there. There was an announcement, they called it the 'modernized traditional applications'. Any take on that and how that fits into the discussion we were just having? >> Well they talked about that a little bit today, not to put in too much of a plug, but we actually had a 45 minute discussion of that with TechFieldDay on Monday and it was embargoed. But the video is actually uploaded now and so if you just Google 'TechFieldDay Docker modernized tradition applications', there's a much deeper dive into that and really what that means and essentially, it's a take on the old P2V strategy that we saw in virtualization that it is possible to literally just scoop up a traditional application and put it in a container. But it's doing more than that and there's all sorts of things that are going on here there identifying which components are part of the application, they're helping you set up the network so that the application will connect still the right way. And I think by choice, Docker didn't really want to emphasize all the real nuts and bolts. I mean, they showed a great, well, an amusing demo of this in action with Ben playing the straight man at the keynote, and that's worth watching as well, but it remains to be seen to what extent they're going to be able to modernize traditional applications and containerize traditional applications. >> Okay, so Stephen, one of the things that is probably the least mature in the Docker ecosystem is storage. I know it's something you've spent some time digging into, what's your take on where we are with storage and containers, where it needs to go, what's the truth and reality? >> Yeah, well my, as you say, my background is storage. And I love storage, I really do. But absolutely, Docker, when I first started experimenting with Docker, I was really blown away by the sort of amateur hour storage approach that they took, I mean, it was essentially, here's a company that knows nothing about storage or networking, building a storage and a networking system. You know, what's wrong with these people? But over time, I've kind of, my view has become a little more nuanced. Because I see that Docker wasn't trying to build an enterprise-grade storage infrastructure, they were trying to build a storage layer that would allow you efficiently to deploy containers. The whole idea always was that storage would be external to the container. And if you're using internal container storage, if you're using the layered file systems, you're doing it wrong if you're doing any kind of real IO. And so, you know we saw a proliferation of plug-ins to allow you to use real storage systems, enterprise storage systems. Ben mentions Nimble and NetApp and companies like that. And in addition, we're starting now to see a whole raft of really interesting, basically container storage arrays. So you've got companies like Storage OS and Portworx developing real enterprise-concept storage specifically targeted at containers. And I think that that's really what's going to happen, is we're going to have the containers using the layered Docker storage but real heavy IO and enterprise applications are either going to us plugged-in enterprise storage or Dockerized enterprise storage. >> Reminds us a lot of what we saw with virtualization- >> Stephen: Absolutely. >> We spent a decade fixing that, I actually remember at Intel Developer Forum, gosh was it like two years ago, Nick Weaver, good friend of ours, works over at Intel, used to work at EMZ, goes to this presentation, I get up at the end, I'm like, 'hey Nick, how are we going to solve all these issues like we did for VMware?" And he was like, 'oh my gosh.' >> And it's pretty much the same story, isn't it? >> It is that same story. >> You know, we're seeing basically the same thing, like virtual storage appliances equals container storage appliances. >> The oversimplified thing of it for me is I felt like we moved along faster with storage and networking took a long time in the virtualization layer, and here it's flipped. Networking seems to move along a little bit faster and storage is there and it's a little nuanced as to what that storage solution looks like, it's not just like, 'oh, we put it all in the hypervisor and eventually it works and we do everything in VM layer.' It's like, well, containers are a little bit different. >> Yeah and some of these container storage solutions are really clever. They've take the lessons from virtualization, from cloud storage, they're building distributed storage, it's really cool. But I think there's another thing to think about there too and that's that Docker invested pretty heavily in creating a, I don't want to say a real enterprise networking layer, but a better networking layer for Swarm. And I think that that may be a road sign of what they may do for storage as well. I think we may see Docker developing a more advanced storage layer, maybe not an enterprise storage layer, but at least something scalable, something distributed for Swarm customers. >> Yeah, I want to get just a little broader from you, just your take on storage these days. I look at adoption of Amazon, VMware's going to go on Amazon. Look, Azure Stack's coming out this summer and you know, we're going to have the S2D as the storage layer for what that's built on. What's the storage market look like from the Foskett viewpoint? >> Well storage is really conservative and when you talk about the market and you talk about the technology, these are two very different things. So the technology is rapidly advancing, we're seeing the world is right now being blown away by the current wave, which is distributed, NVMe, ultra high-performance flash storage, exemplified by a company like Excelero, for example. That's absolutely the coolest stuff out there right now. But then the market is still adopting SAN. You know what I mean? The market is still, you know, 'hey, should we implement iSCSI? Hey, should we look at NFSv4?' Things like that and it's a real kind of facepalm thing because you look at the reality of storage and it doesn't keep up with the promise of enterprise storage. But it's, yeah and then there's the whole aspect of sort of cloud storage, off-premises storage and that is also a potentially game-changer for the market. But overall, I would say that you'd be a fool to bet on radical transformation of storage. It's just not going to happen. You know, that's why HP's going to get tremendous value out of buying Nimble. That's why NetApp and Dell EMC are going to be selling a lot of product for a long time. Because although they're innovating and advancing and keeping up with some of these new waves of storage, the truth is most buyers are buying very calm, boring stuff still. >> Alright, Stephen, unfortunately we're running low on time. Why don't you be the final word, let's talk about the community aspect. I loved, you come into a lot of these open-source shows, it's just got a great vibe, enthusiastic, really people that want to learn. And I know that always excites me, it's the kind of thing that you love, hanging out with those people too. What's your take on kind of the Docker ecosystem and community? >> It's wonderful. I mean, it reminds me of how VMware was back, well, the last decade. It's a warm, inviting, exciting community. And one of the things that I really want to highlight here at DockerCon that I've seen is that it's a lot more of a diverse community than I've seen traditionally in IT. I'm more of in enterprise IT and so there's a lot of people walking around that look like me. And looking here, there's a lot of people that don't. And that is fantastic. Docker has done a great job of emphasizing diversity, they've got onsite child care, they've got, I mean, Solomon tweeted that there's 20% women attendees at DockerCon. To me, yeah, the vibe is great, but wow! Talk about broadening IT and talk about modernizing IT. That's modernizing IT. >> Alright, well Stephen Foskett, always great to catch up with you. I'm sure I will see you at many conferences throughout our travels throughout the year and we've got a full day of coverage here from DockerCon 2017. Solomon Hykes is coming on, we do have Visa who did the case study, many other partners, Oracle who made an announcement today. I've got a couple of service providers who actually participated in Stephen's TFDX event here before the event. So stay tuned for all our coverage and thank you for watching The Cube. (techno music)