 We're back live in VMworld 2012. I'm John Furrier, the founder of SiliconANGLE.com. This is SiliconANGLE.tv's theCUBE, our flagship telecaster to go out to the event, extract the signal from the noise, and keep your eyes out next month when we launch SiliconANGLE.tv within broadcast studios. You're going to see that starting next month. But I'm pleased to have on a panel here about Cloud Foundry, James Waters, who is a SiliconANGLE alumni and now the senior director of Mark programs. What's your title? Yeah, I'm responsible for ecosystem development for Cloud Foundry. Ecosystem development for Cloud Foundry. James is a distinguished member of the Cloud Arati influencer, and just overall, great Cloud Mojo. Great to be here, John. James, welcome to theCUBE. Lucas Carlson with the App Fog welcome as well. Thank you so much. So Lucas, let's start with you. You're actually out there in the trenches doing the work. Give us a quick update on what's going on in the cloud because a couple of things that we've been talking about at SiliconANGLE is this emergence of data infrastructure where Flash has really changed the game in the past 24 months with just benefits that just are all over the place in many use cases. Caching layer, scale out transactions, scale out NAS, scale out analytics now with big data, and obviously with big data, on top of that, you see in the storage business become really kind of upside down. What that's done is it's created an on-premise, it created a stall on the mass migration to cloud and created a little bit more head room or time on the on-premise deployments. People could get the economics and they can get the performance with Flash, so it kind of stalled the mass exodus to the cloud. Although cloud, there'll still be some migration, so hybrid cloud's hot, so just help us understand on the trenches what's the current marketplace for the cloud landscape with obviously public, hybrid, and private? Yeah, what we keep seeing over and over is more and more mobile web and social applications being built within these enterprises, especially in these lines of business, we see increased development needs, and what we're seeing is that all of these different technologies are trying to be incorporated, and we're seeing different technologies like Node, and Ruby, and Python increasingly being put onto the cloud. And this is where we've seen an emergence of things like cloud foundry being an important and critical piece to tying clouds together, bringing together public and private, and a hybrid solution. Could you talk about, and James, you can chime in on this one too, obviously. This has been a massive excitement of public cloud going way back to when it all started. Obviously, we went with Amazon as the use case, Rackspace, real tooled up pretty quickly. We saw that move very fast. Now you got OpenStack out there, self-defined data center, all this stuff's going on. Talk about specifically around the use cases around how big data has changed the cloud equation, and specifically DevOps. DevOps, obviously, developers who actually do operations, Facebook, Twitter, and they have one organization. How has the DevOps world changed with the changing adoption or demand of cloud? Yeah, I mean, I think the number one thing that DevOps is a part of a larger trend is that how you operate at cloud scale is going to be a big part of how you differentiate and think about yourself in terms of your organizational competencies. It's not enough to run in a siloed organization with, hey, you've got one app, we've got one guy that runs that app, et cetera. That was the old model, cloud got rid of that. One of the interesting things that Lucas has done is he's talked about the difference between DevOps and NoOps, and in the PaaS world, we like to think that someone who's consuming a PaaS, at least as how Lucas presents it, actually just pushes an application to an abstract place that they don't actually understand all that DevOps underneath it, and that's a really important distinction that Lucas has made a lot of heat with. Yeah, one of the things that we've noticed over and over is DevOps has been one of the solutions that developers have brought into the enterprises in order to try to solve these problems. But what we've seen is that these developers, especially as they go from 20, 40, 60 to 1,000 developers within these lines of business, they each have their own DevOps. They have their different approaches, and that's where Platform as a Service can step in and say, here's a way to standardize. Here's a way to approach the problem in a way that all of your developers, 1,000 developers can be using it and deploying in the same way. First of all, we love DevOps. For the folks out there, we actually have a dedicated site called devopsangle.com where we're really focusing on the developer community. It's become kind of our developer angle and which might morph into a much larger category. But the things like Rails, Heroku, we've seen the great stuff happen with the rapid cloud agile launching programs and Node, obviously it's another one. We did the Node Summit, we had theCUBE at Node Summit. All great stuff, IO centric and low-cost high-performance capabilities on demand, all the greatness of the cloud. But big data has been an emerging disruptive force with that because people want analytics as part of their solution. You seeing anything around how that big data is affecting the past world? Yeah, so big data has been one of those things that in order to consume it, you need big applications to handle those big data and that's also where Paz stands out. One of the things you can have developers trying to approach the DevOps solution for big data. But what Paz has in the box is the ability to scale out a system that can consume big data over thousands of instances very quickly. So we've been talking about us, it's on-prem dynamic with Flash. It seems to be, yeah, with legacy stuff, it kind of stays there until they really totally get beyond the point of, okay, let's go to the good cloud. All new use cases are going with cloud. So can you share, Lucas, what you're seeing in terms of like just run down the list, top use cases that people are using the past clouds for? Yeah, top use cases are very easy. Mobile applications, social applications, and consumer web. These things are dominating the transition to the cloud and it really is one of those things that you see more and more applications. It's really been a 10-year trend. 10 years ago, if you were a developer, you couldn't get a virtual machine outside of Core IT. Now you have 10 different places where you can rent virtual machines and in fact, you're encouraged to. And that has really led to a wide variety of. Just a speed of deployment and just a ramp up of clusters at Google IO, we saw a map bar, spin up massive benchmark. This is so awesome, I mean, you can't deny, and that's obvious, right? So the question I want to ask you guys is really something that has been kicked around as commentary in the word of mouth networks around, platform as a service is a race to zero. So people say it's a race to zero. They don't want to go back to the hosting model where it's just driving down, getting lower, lower cheap kind of positioning. Just share your commentary on that. I mean, I've been saying that, okay, that could be one scenario if you look at it that way, but we're seeing business models that wrap services around what looks like a race to zero. So just what's your perspective on that and can you share any color around those things that are involved in past that doesn't make it a race to zero? The one thing that people like to do sometimes, I think, is exclude the human factor from the equation. So they say, hey, here's how much this hard, tangible, good cost, and what are you adding on top of that? What, how thick is that layer? The human costs of our current deployment and operational models are immense, just immense. And you can talk to some of the Cloud Foundry early enterprise users who think they can get 50% savings from using Cloud Foundry. And that's not about making the hardware more efficient per se or doing any of these traditional optimizations of just the physical stack and the software. That's about the organizational change that these concepts drive. DevOps, PAS are gonna unleash 50, 60% savings and operation or more, and they're also gonna attack agility in a way that no one has, and this is a pretty common theme. But the reason is, John, when you think about traditional IT, it was built to deploy process-centric ERP. So if it took you 18 months to roll it out and there were a lot of meetings, that's fine, because you had to make sure you closed your books, you didn't go out of business. That's a different philosophy behind IT, and that's what built the existing data centers where every app had its own team. The difference now is you're trying to go after a mobile opportunity that might last six weeks. If you don't get your mobile app out in the next six weeks, you're gonna be in the bottom of the list. If you don't have marketing for your next conference that's ready and on message, as you talk about, right? Message only lasts so long in the market. People are trying to execute real investments in two, three, four month time scales. It's completely different. I think where I like is the agile message, the people that I hear that comment from are either old school cloud guys, when cloud was still a lot of bare metal and they had to build their own clouds and it was still slower. But they don't get this notion of rapid deployment, and the cost structure's so low. We've seen a similar argument with Flash. So when Flash at the scene was like, oh, Flash is really, really expensive. Well, yeah, it is expensive. She's not gonna see it in all the little consumer devices, but when you look at the cost of a Flash implementation and the reduction of a million dollar sand at EMC, about to buy from EMC, hello, that's disruptive. So the labor cost or the cost of ownership or whatever you want to call it, total cost of ownership is a factor. Okay, I don't think there's any debate on that. I'm gonna put the stake in the ground and say that's a fact. So I don't think it's a race to zero. I think it's definitely an enabling opportunity. So I want to just get with you guys, what are those enabling things that are going on around past that are gonna create opportunities and disruption? Well, the thing that's been slowing down development is traditionally when you create applications, it can really take six weeks just to get your ops guys to deploy the code. And if you have an update, if there's a problem with that code, it could take another six weeks. And those kind of timeframes aren't possible to sustain anymore. Now we need two, four weeks to deploy a full application, not just waiting for ops team to get around to deploying your code. And that rapid development is exactly what Paz enables. What's the biggest misconception of Paz, Lucas, in your opinion? The biggest misconception of Paz is that it is just a feature on top of infrastructure. It is not yet another sort of resource. It's actually a different way of thinking. When you're deploying and managing apps with services and services have to talk to those applications and knowing how to scale each of those independently, that's not just another kind of resource. And it's a layer that is very thick on top of the infrastructure. So we're going to talk with Chris Hoff after this about security. So James and Lucas, what's your take on security, the security question around Paz? Where are we? What ending of the game is it? How reliable? And when I say security, I mean, comes in, you know, obviously intrusion detection, I'm intrusion and also data security. I mean, I'm not a security expert. So I don't, you know, Chris is standing over here taking pictures of us doing a little Irish jig. You can ask him all about security. Don't be nervous, come on. What I would say is that Paz lets you bake in a lot of things automatically, whereas a normal developer working alone out on Amazon is setting a bunch of configuration files himself. And Paz lets you do that systematically and it takes decisions away from individuals that they don't need to make. And so I wouldn't want to say where we are exactly on the continuum, but I think there's a huge opportunity to standardize. We're stepping in the right direction. Security is one of those things that has been traditionally done very well with operations teams. And Paz brings that control and governance back into the operations teams where it wasn't before with DevOps in the cloud where developers were trying to do it on their own. So obviously, Jane's on the ecosystem side of VMware, so I'm going to ask him the VMware question. Sure. Lucas, I want to ask you more of the entrepreneurial question. You guys are growing, what's the trajectory for Paz? For not just you guys, the industry, how do you see it morphing? Where's it going to be flying to? The trajectory is exponential. A year ago, we had about 10,000 developers using the service. A month ago, we had 40,000. Today, we have 60,000. It's growing like a weed. And it's really this developer bottom-up approach. Developers want this and that's driving everything. What's the number one thing developers are looking for? Obviously, convenience? Is it speed? Is it agile? All of the above? It's to get their applications deployed faster and easier than ever before and then scale them. There was a ton of non-value work that was being done and that's the easiest way you can say it is there's a ton of non-value ad work that was being done where an individual had to turn to a team that would respond in their timeframe to help them do that non-value work. That's going away with Paz. Awesome. James and Lucas, now let's talk about your companies. Obviously, James, you've had an interesting journey at VMware. I remember when you started, they were kind of kicking this around. There's a lot of cloud movement with Cloud Foundry. It was pretty obvious they needed to have a cloud strategy with where VMware was going. Now we're hearing about multiple clouds from Steve Herrod. How does Cloud Foundry go forward? First, give us the update on Cloud Foundry and then the vision of multiple clouds and software-defined networking fits into that. What's the angle there? So my angle at VMware is Cloud Foundry ecosystem. So my vision from day one was that this had to be multi-cloud. So you hear Steve talk about multi-cloud, the opportunity that we approached Lucas with to work on together was all about multi-cloud. And I would say that Lucas has taken multi-cloud even further. He now has deploys. And how many clouds can you play with? We have six different data centers and four different vendors. So we've taken what used to be, the whole world was one little cloud hosting area and we've taken Paz and dropped it everywhere. And that's really, really important. When a major enterprise customer is talking at DeployCon and why they chose Cloud Foundry was all about multi-cloud. They said, hey, we actually, we've tried these other little hosted passes and then we liked the idea but we could never bring that in-house. And we need something that we can use both hosted and in-house. And it's really an immature market in some standpoint with a lot of people doing those single non-multi-cloud deployments. And it's by getting the multi-cloud, it's by getting the opportunity and choice to bring it wherever you want that we're really going to unleash this as an economic force. We have large PO's being cut, your organization is changing. The kind of things that people have bottled up in the vision, those are going to start to happen. I got to jump in and do a quick follow-up question. So Paul Moritz is moving on to sit and work on the next canvas for VMware. Pat Gelsing is going to take over and run the operational, run the trains and get the house in order. What is Paul going to work on? I know you don't have insight in this but from James Waters' perspective, being the strategic genius that you are and the Corp Dev side, what is he going to work on? I mean just from a high level market perspective, he's going to work on creating more software elements in the cloud. I am, huge respect for Paul. He's one of the reasons I came to VMware. Every time I get a chance to interact with him, it's a total pleasure. What do you think he's going to work on? I have no absolute comment on any management decisions. Okay, Lucas, what do you think Paul Moritz is going to be working on as he paints the next canvas for the future of VMware and EMC, might I add? Yeah, well obviously I can't speak for him but if I were in his shoes, I would see platform as a service as a very important next generation for the cloud that would sink your teeth into. I mean I think he's going to work on the software defined data center vision which is going to be essentially a very seamless, abstracted away operating environment for computing, mobile and big data. So that's a hard challenge. I think it's going to be some science involved and more M&A activities. So I think that's, I mean I'm psyched for him. I think it's going to be great and he's moving to right down the street from me so I'm going to be able to pop in and I'm going to get an interview with him. He has yet to be on the queue by the way. Any news on you guys recent acquisition? You want to announce, share that with the... Yeah, we're just announcing this morning that we acquired Nodester, yet another node Paz that is actually a market leader for nodes. So now we can see the bringing together of different technologies. So that would be safe to say that you buy into the SiliconANGLE Wikibon vision of IO centric infrastructure. Yeah. Yes, of course, IO node is great. So what's next for you guys? What's on your plan this year? Our plan is really to help enterprises get their cloud strategy together. Brought bridge the public and private clouds together in a way that looks the same and is able to enable core IT to be as a service to its lines of business. What about the infrastructure as a service market? How do you guys see that one? Yeah, so we have another partner in the ecosystem Tier 3 and they have a blended infrastructure as a service and platform as service offering. And that's been a very nice bridge for a lot of enterprises that are adopting this. They've got, and Lucis has the same capability because he's on Amazon, you can do the same thing there. But I do think that's been an important part of the adoption for enterprises. Hey, some of this is going to run infrastructure service, some of this on platforms of service. It's been key, but more and more is going to come into the Paz abstraction over time. You're going to say, why am I configuring all of these VMs to run applications when I can just deploy an application to a platform? The one thing I'd say about this is, I get excited in the morning when I wake up and I read a new customer use case. So I've got lots of ideas, happy to share them on a blog or tweet, et cetera. But what really gets me going is someone like Intel who's here talking and they're saying, look, we used VMware to take our deployment time from six months for a server to a couple hours for a VM, even less. And that was awesome. But we need Paz to get us from an idea to production in a day with just the individual. And that's something where you can really see the difference between what an infrastructure service can provide and what a platform as a service can provide. Because with the platform as a service, they can have a couple developers working alone, push their code, be in production if they want with all the processes and everything glued up by the platform underneath it. And they're here talking about that. That's not coming from marketing or from us. That's a quote from them, idea to production in a day. Yeah, I mean, to me, this is phenomenal trend. Just got to bake out and deal with the flow of the market. I mean, it's a lot of demand. So with that, I'll ask my final question for you guys if you all can give commentary on it. Two parts to the question. Part one, what happened in the past 18 months that really surprised you? Like, good or bad? Like, wow, I don't think that's gonna happen or it could be anything, acquisition, trend. And then two, where do you see in the next 18 months disruption really happening? Like real, really critical disruption that in a good way that creates opportunities and new entrepreneurial ventures or wealth? Go ahead, Louise. In the last 18 months, I think one of the things that really surprised me was not seeing platform as a service yet being adopted. It's been one of these hot topic things that everybody talks about. And I think that some of the things that held it back were the immaturity of some of the pieces. And I think we're now seeing the opportunity. So looking forward 18 months, I think seeing enterprises adopt private cloud platform as a service from vendors like VMware and AppFog is really gonna be what enables the next generation of developers to build great companies. So developer disruption. I think so. Okay, cool. I think, you know, just speaking for me personally, the number one thing that surprised me is how people have underestimated that you're gonna continue to have data centers that are privately operated. And there's been this huge confusion like, oh, it's all cloud, it's Salesforce talking, and hey, it's all hosted CRM, that's the future. And I don't think that's the case. And so people have underestimated how much software-defined data centers and how much fully automated data centers are gonna really rock the economic house. And they've overpivoted towards, oh, just hosting solves all. And they've underestimated how much amazing software engineering is still yet to come to operate at large scale with high automation. And people have been really confused by what is cloud, what does that mean? And they've underestimated that. So that's where I'm laying my bets. That's why I'm working on the hybrid or multi-cloud technology of cloud foundry at VMware. That's why I'm, you know, that's making my bet there. Yeah, I mean, Om Malik just wrote a post on GigaOM, talking about Evernote, kind of slamming box.net, if you will, but people, and they have a consumer product, but they're trying to get in the enterprise. People don't realize how hard it is. I mean, it's not, it's hard. To have a standalone cloud product for the enterprise is really difficult. And you've seen that, you know, Amazon's done a good job because, you know, on the high end, if they gear up, they can do that. But it's hard in the enterprise and cloud foundry is at the beginning of its journey in growing. James Waters, charge ecosystem, Lucas Carlson with app fog. Thanks for coming on the queue, appreciate it. We'll be right back after this break with our next guest. Awesome to be here, thanks. Thank you.