 Live from Vancouver, Canada, it's theCUBE at OpenStack Summit Vancouver 2015. Brought to you by headline sponsors EMC and Joypling by Red Hat and Cisco with additional sponsorship by Brocade and HP. And now your host, Stu Miniman. Welcome back to theCUBE, I'm Stu Miniman with wikibon.com. Here are SiliconANGLE TVs feature program at the OpenStack Summit in Vancouver for 2015. Joining me for this segment, we're joined with Radesh Balakrishnan, who's the general manager of OpenStack at Red Hat. Welcome back to the program. My pleasure. All right, and we've got Mike Cohen, who's the director of product management with Cisco. Mike, welcome to the program. Thanks. All right guys, so, you know, OpenStack, it's every six months we get together, talk about what's going on, you know, talking about the maturity of the products, you know, two companies that, you know, have been in industry, helped drive IT forward for years. Radesh, you know, you've got OpenStack in your title, what's your take on the show so far, and you know, what's the state of OpenStack? It's a great place, you know, part of the course is what I would characterize it as, as you rightly called out, the maturity of the product and set of ecosystem around it has really come through in the last couple of years, if you will. We're also, from our vantage point, seeing real customers deploying it in production across both the enterprise as well as Telco, and clearly partnerships such as one that we have with Cisco, which is across multiple divisions, we're going to focus on ACI in today's discussion, but we are very, very excited about the fact that the number one provider of year for cloud out there, as well as the number one open source software provider that's us are getting together to address a challenge and an opportunity. All right, so Mike, you know, since Cisco's had a good presence in this community, one of my favorite interviews from last year was actually Lutaker talking about how open source is really trying to work its way inside of many parts of Cisco. Talk a little bit about where things are with OpenStack and Cisco. Sure, absolutely. So Cisco's a major contributor to OpenStack, and honestly doing upstream contribution is one of the primary focuses of a number of our groups, including mine on the application-centric infrastructure side. Really, our goal is to partially enable us to sell some of our products and switches and servers, et cetera, but also make the community better and make OpenStack better so that we can help our customers build larger deployments and actually accelerate adoption of the platform, which also helps us from a business perspective too. All right, so I love both of you talked about the customers in the first segment, which is really our maturity of the environment, is, you know, what are customers doing? What are they asking for, and how are they doing it? Like, if I could start with you, I mean, you talk about things like, you know, your servers and your switches and everything. How much is OpenStack part of the conversation today from the user base? Sure, so actually, OpenStack's becoming a really major part of the conversation, and I think, you know, if you go back to two years, that wasn't as much the case, but now we're really seeing it as, you know, really a primary ask in almost every one of our major accounts, that everyone is building an OpenStack cloud, and they're actually looking for a really stable environment on which to do that. And one of the things we did with our ACI product was actually build essentially an SDN solution that was designed around policy to accelerate application deployment, and do it in a performant way, in a very scalable way that can actually act as a back end for platforms like OpenStack. All right, so Radesh, you know, of course Red Hat's been involved in OpenSource for a long time. Maybe talk a little bit about, right, how's the discussion changed, you know, is OpenSource maybe not even on the table when we're discussing it? You know, where does OpenStack fit into the discussion for Red Hat? It's an excellent question if I kind of layer my thoughts. At the highest level, if you look back, let's say even seven or eight years ago to today, there's no question about if OpenSource, right? So OpenSource definitely has a role in the next generation data centers. There's no doubt about that. The discussions are very mature in that. How do I bring in OpenSource-based solutions with the same dependability and enterprise class, supportability, et cetera, that customers are used to, right? So that's a very maturing overall umbrella around OpenSource itself. Specific to OpenStack, what we are seeing is that, you know, two years ago when we were just having OpenStack alone conversation, a lot of customers were, you know, especially the CTO office calling and saying, tell us about where you're taking OpenStack, how can we work together, et cetera. Two, OpenStack being a foundation for a full stack of solutions. So we're seeing every other opportunity that we're engaging in, SF as a storage solution being dragged through, cloud management, which we have cloud forms as an offering, increasingly becoming, you know, every other conversation involves that as well. And then last but not the least, we're also seeing customers changing away from the vernacular of IAS and PASS and SAS. You know, I was joking with somebody, IAS and PASS have become PASS-A now. You know, they're viewing it in conjunction, right? I think it's happened to a large extent in the public cloud as well. And we are seeing a lot of open shift on OpenStack, which is our PASS on IAS convergence also happening. So whichever way you look at it, the level of sophistication of questions to discussions to actual deployment plans have all matured tremendously in the last two years, which is heartening given where we are. All right, so maybe, guys, can we pull kind of ACI into the discussion? You know, ACI has been a major initiative at Cisco. Now, you know, put together pieces from ACI, OpenStack, and the Red to Hat Cisco partnership. Sure, absolutely. So, you know, obviously with ACI, what we tried to do was build an SDN solution that tied together, you know, an overlay and underlay environment and essentially did it with this application-centric policy model. You know, one where you could actually use policy to describe your applications and deploy them in an accelerated manner, which actually acts as an excellent backend for an OpenStack environment. We're actually want to build a cloud and I want to be able to place VMs anywhere and deploy applications incredibly quickly. We're seeing a lot of exciting new projects come on to enable that kind of development. And ACI can actually act as the backend across all those systems. And Red Hat actually becomes a critical partner here because honestly, when we go to market at Cisco, we're bringing our ACI solution. We need to partner with essentially a company that can offer enterprise class grade of OpenStack that our customers can deploy and stand off and actually achieve the application environment that they need. All right, so Redesh, we know Cisco does their great CVDs to help put together the solution. What's Red Hat doing to help bake out the solution, make it easier for customers? Yeah, it's definitely working on CVDs, et cetera. Before we went there, kind of back to your question about how do we see the solutions coming together for customers. If you look at a classical OpenStack evaluation, a key pain area that customer surface is around networking in general, right? So clearly, that highlights the fact that a lot can be done in terms of making things better. Another reality also is that all the infrastructure constituents, even within any large organization, is now realizing that developers are becoming the king. Application centricity is a reality that's in front of all of us. So you tie these two dots, ACI fits in nicely because it takes the complexity from the hardware layer to an application-centric policy-driven framework, right? So that's why there's natural synergy in terms of our solutions coming together to be able to address customers' problems. The other reality also is that to this point, and it's an SDN solution. So we've seen how with virtualization compute becoming pervasively virtualized, et cetera, networking has been the bottleneck. Now, clearly, SDN solutions provide a solution. So if you can address that challenge as well as make it application-centric, that's an amazing place for customers to get to. So that's how we see the synergy. Now back to your question about what are we delivering in the market to help our customers. Of course, we are doing the joint development of the solutions itself so that really, really, the equipment or the software into a customer environment, they're not testing it for the first time. It's a certified, pre-tested solution from both the parties. The second is creating artifacts such as CVD, et cetera, so that the deployment can be very predictable, et cetera. So all the things you would classically expect around a line, go-to-market are what we're doing. All right. Is there something you want to follow up on that? No, I was just going to say, one of the other areas where we're actually working together with Red Hat on, and I think we add a lot of value is on the operations side. So one of the areas, the OpenSec has done a great job at building a cloud environment. And I think one of the areas where Red Hat adds value and where we're adding value with ACI is actually making the whole thing operational and allowing essentially all of the operational infrastructure teams to manage the cloud and actually understand what to do when something goes wrong. And this is where vendors like us can actually add a lot of value to these environments. Yeah, actually, I loved in the keynote this morning, Mark Collier said, what do we call OpenStack? Because is it a product? Is it a cloud? Is it there? And he said, we're an integration engine. So we talked about how containers, take Docker swarm or Kubernetes and plug into that and then heat underneath puts it. It really helped to kind of summarize for me where it fits in because you guys both have hardware and software pieces that you have involved. So OpenStack is kind of that piece. Yeah, I guess any comments on that? Yeah, no, I think we're definitely seeing a little bit of a shift in the model in that originally obviously consuming a VM and being able to achieve VMs on demand was a big step for enterprises and we saw a lot of people adopting that. And to some degree, that's happened, right? We've delivered that. Everyone's happy and there's clouds being built and many places are already deployed. The next step beyond that is moving towards these application-centric models and services architectures where they can actually use things like OpenShift and actually start thinking a layer above in terms of how do I deploy applications? How can I use platform layers or layers in built-in containers to think about my applications and design them in a better way? And I think that's what's happening and that's an exciting step in the community right now. Something we're enabling at the lower levels with ACI and the Red Hat's doing a ton of interesting work around it too. Yep, I agree. You know, that's a massive opportunity in front of us. You mentioned container, right? So who is not talking about containers? And we believe that from a Red Hat perspective, we uniquely position that we can have a, if you want to go for a, I want to string it all together on my own. You have Relatomic as an offering which we made available a couple of months ago to a full-fledged finished DevOps environment with OpenShift running on OpenStack. That's the other option we make available too. I think the whole shift towards application centricity, microservices, all those are amazing opportunity for us to go, you know, re-cement our value proposition in the eyes of the enterprise customer. Yeah, no, Radesh, actually, I called out to the community a bunch of times. Red Hat actually delayed the release of, you know, the last Red Hat Enterprise Linux by a few months to make sure that Docker was the first class citizen. I mean, talk about, you know, commitment to this early and really getting it, you know, so kudos for that. Mike, so I won't, you know, I'm going to bring up a term that I actually hadn't heard at all this week. That software-defined meme that's out there. I like when we're here at OpenStack, you know, we're all talking about software. I mean, we're talking about open source, we're talking about software. In some other conferences, it's software-defined this or that. What's your take on where we are with this? You know, what separates what Cisco's doing from, you know, some of the other pieces that kind of call it software-defined something? Sure, absolutely. And I think, you know, this is definitely something that's happened in the software-defined networking world that has come to mean all things to all people. And honestly, the clear definition of it is now morphed to a point where it's hard for people to articulate exactly what you're trying to achieve. And, you know, one of the things that I like about what we're doing with ACI is that we actually try to drill down to a concrete problem that we know our customers have, which is they want a fast, agile, performant network that allows them to accelerate application deployment. They have an application, they want a quick way of deploying that in a very flexible way across their network. You know, in a data center environment, that's the concrete problem we saw that is the biggest pain point. And that's what we set out to solve. You know, and honestly, when you, and that's the ACI solution came out of, you know, that problem statement. Now, in my interpretation of software-defined, you know, software-defined networking, I really map it to that problem space and try to map it back to what our users are having pain around, which is how can I launch applications, how can I do it fast? You know, a lot of the stuff around control planes and how they fit and thing, open flow, I think they're red herrings, honestly, in the larger sense of what people are trying to achieve. Yeah, so from our perspective, you know, it's beautiful that everything is moving to a software, right? But, you know, clearly, as the examples he pointed out, we have to be able to, you know, leverage all the assets that are the hardware layer as well intelligently. It doesn't mean that everything becomes commoditized at that layer too. So, that's where I think partnerships, such as one we have with Cisco, we talked about ACI a lot. The same applies to UCS, for example, or how the inter-cloud offering of Cisco itself is going to be structured, et cetera, right? So, I think, BC is a huge opportunity. There's a tectonic shift. Everybody's acknowledging that. Open Source has a much more prominent role to play, and our belief is that by betting the farm in conjunction with broader shoulders such as Cisco, we can reach broader set of masses and deliver on the promise of software to find anything, right? Yeah, so, Radesh, you know, it's got to be funny being at Red Hat in a time like this, right? It's like, wait, it's a software-based, it's open source. You know, we've talked to a lot of people. It's like, how do you monetize something like this? It really comes back to, you know, why is Red Hat positioned to really take this opportunity of OpenStack and help bring it to the masses? Right. So, you know, from our monetization model is always being subscription-based. So, you know, let's take the case of OpenStack. Every six months, there is an upstream release. What we do is we take about a couple of months to enterprise hard in it. That includes bug fixes, which are obvious, but also back-porting features, because at any point in time, we are working towards the next release as well. And then the third one is we have deep product management level engagement with customers and telcos and service providers, et cetera, that makes it into upstream as well. As you already know, we don't do anything which is shadow code trade, right? Everything we do is upstream first. So, it's these value elements that customers get by going for a subscription of Red Hat OpenStack platform, which is our offering. One point worth highlighting is that one of the conscious choices we've made is to architect our OpenStack offering with our Red Hat Enterprise Linux offering. And people ask, why? That has to primarily do with the fact that if you want a functionality set for standing up a cloud and have a life cycle thereof, you need to engineer the components together. KBM layer, the operating system layer, and OpenStack. So, we have seen that both from a rearview mirror of where the support calls are coming, as well as forward-looking feature innovation, et cetera, our ability to influence all these three layers and package it together is valued by our customers as well. All right, so, we're running low in time. Mike, I want to start with you, you know, final thoughts. You know, either something cool you've seen at the show or something that, you know, Cisco's helping to kind of stand out from the crowd that you'd want to point out. Sure, so one of the things I would call out, you know, that we're working on, and we're working on conjunction with Red Hat is group-based policy. So, this is one of the ideas we're driving forward, where we've taken some of the ideas that had a genesis with ACI where we were thinking about different ways of describing policy. We're actually making that into a project that we can run on top of OpenStack with any network backend, not just with our solution. And, you know, part of the idea there is to actually bring these ideas of thinking about applications differently and tying them together in terms of services and make it easier to use it for the entire community. All right, Redesh, the same question to you. So, from my perspective, the fact that the show is becoming more boring and enterprise-y, I love it, right, weirdly, because, you know, that's what we, that's the set of customers that we engage with, that's the customer mindset we understand as well, and to some of the points that we were talking about earlier, I think it has reached a stage where, you know, the crossing the castle moment has already happened. It's gotten beyond the innovators in terms of early adopters, in terms of adoption, it's becoming more mainstream. So, to be able to have all these rich set of conversations with partners as well as customers to help shape the next version of OpenStack, that's an amazing place to be in. So, you know, my twice a year pilgrimage check mark, happy customer. All right, well, gentlemen, thank you so much for sharing with us and the customer stories, the partnership that you're growing and the efforts going on OpenStack. Got lots more coverage here from theCUBE over the next day and a half. Stay tuned, watch all the replays up on YouTube.com slash SiliconANGLE, and of course, all the live programs and lots of upcoming events, including the Red Hat Summit, you know, just a month from now back in Boston on SiliconANGLE.tv. Thank you very much.