 Live from San Francisco, extracting the signal from the noise, it's The Cube, covering Oracle Open World 2015 from Studio C, brought to you by Cisco. Now your host, Stu Miniman and Jeff Brick. Hey everybody, you are watching The Cube. We are live at Oracle Open World 2015. We're on the exhibit floor in Moscone, South. A lot of action, a lot of video, but this is The Cube. We go out to the events, extract the signal from the noise. I think this is about event number 70. Busy fall tour. We're really excited to be here. Joining us next segment by Stu Miniman from Wikibon and a first-time, first-time guy at The Cube, Marcus Phipps, senior manager, data center, cloud marketing from Cisco. Welcome. Thanks. Glad to be here. Absolutely. So welcome to The Cube. So you are all about UCS. It tells me in my notes, you're the deep dive guy. I'm one of the deep dive guys, yeah. So we'll try not to go too terribly deep, but what I wanted to talk about today is the concept of UCS integrated infrastructures and really what we bring to the table there. Alrighty. So what exactly does that mean then? Okay. So what converged infrastructures or integrated infrastructures mean is it's basically taking computing, networking and storage and integrating them together into a predefined solution. And really when you look at where customers have been and really what their key challenges are, it really comes down to two things. Complexity and speed. Technology is hard. It's difficult. If you're trying to integrate all of this yourself, it can be really complex and it can take a lot of time. And the other piece that's really important is speed. How fast can I get a service up and running? If I have to take months to build out the infrastructure. That's not flying. That's not flying, right? You simply can't do it anymore and that's putting IT under tremendous pressure. Right. Yeah. Marcus, you know, we've tracked this phase for many years and I lived in an interoperability lab for six years during my career. So, you know, people tend to oversimplify things. They're like, well, first of all, I've got standards, right? I mean, you know, I take these plugs, they go together. It should all work together. And I should be able to take any application and throw it on, you know, any infrastructure, you know, you know, where's kind of the hard work and what is Cisco doing to really simplify these? So you can always do that. You know, we always tell customers, if you want to assemble it yourself, you can absolutely do that. But what's happening is, yeah, I mean, the plugs are going to work and frankly, you know, an Oracle database is going to run on UCS. The challenge usually isn't there. The challenge is in how all of the integration of the systems come together, right? What's the best way of connecting UCS to storage? What's the best way of implementing software to find networking over the entire infrastructure and then deeper into the networking infrastructure? How do you do that? There's probably about 10 different ways you could do it. No customer is going to want to do each of the 10 different ways to figure out what the best way is. So they're really looking for Cisco and our storage partners and our application partners like Oracle to figure out what's the best way of doing that. So I can get that up and running as quickly as possible. So when you're trying to figure out the best way, I mean, what are some of the factors that determine the best way? Because it's horses for courses. It's application specific. It's what do you need? So what are some of the things that you guys look at when you're trying to optimize and figure out the best way? So what we'll do first and foremost is we'll take the core systems that you see of the integrated infrastructure and there's really three of them that we bring to the table. There's the computing component, which is UCS. There's the Nexus component, which also includes application centric infrastructure for software-defined networking. And then there's the orchestration piece. So at the first level, customers are going to want to understand, okay, well, how do those pieces work together? What should my configuration of UCS be if I'm running an Oracle database or I'm running a business suite? What does my service profile look like? Are you familiar with the concept of service profiles? Yep, yep. So for the benefit of the customers watching this, you basically define a server in software, apply it to the underlying infrastructure, and it just works. So there's different ways you can configure that service profile. Here's the best way of doing it, for a database, for e-business suite, and so forth, so that when you're applying that to the underlying infrastructure, you know that it's just going to work. Why? Because we've tested it, we've validated it, we've taken a lot of hard work and made it work. These are your validated designs that have come up. These are the Cisco validated designs that you can find on DesignZone.com, yeah. Okay. So Marcus, we've actually done quite a bit of research on this and kind of the further up the stack you go, it kind of gets exponentially more important as to those integrations. Because even if I've got the infrastructure all set, you know, when I set my applications on top of it, I need to know how that interaction, how the performance has worked, how I balance everything. So I guess the question, you know, we're here at Oracle, you know, why shouldn't all customers just buy an entire red stack where everything from the silicon, all the way up to the application itself, is owned by one vendor? It's, you know, Oracle, you know, doing well in that business. You know, that full integrated stack versus, you know, Cisco, of course, works with lots of applications including Oracle. So we found that what customers were really looking for is a tremendous degree of flexibility. You know, they have existing processes, they have existing vendors, they have in many cases UCS for a lot of their high performance applications or frankly any application. And so that flexibility in terms of infrastructure and being able to tweak things as necessary to bring in, you know, EMC as part of a V specs or NetApp storage as part of a FlexPod or IBM storage as part of a versus stack and then have the ability to configure that as they need. That's why we've really seen the UCS integrated infrastructure construct take off. In fact, according to IDC, I think Cisco had about participated in about 47% of the integrated infrastructures with V block and with FlexPod alone. And we're actually over 50% because EMC is tracked separately but we actually participate in their V specs program as well. So in the integrated infrastructure or converged infrastructure category, we're seeing a lot of success because customers want that kind of flexibility. And even kind of breaking it down a little bit further, you know, you do have the reference architecture piece which is what FlexPod and versus stack and it's the new solutions we're working on with Nimble are today. And then you have the, you know, the kind of preconfigured ship from the loading dock or I should say from the factory floor to your loading dock. And that's what VCE is delivering which even more tightly integrates UCS and EMC storage and VMware virtualization. So maybe could you talk a little bit about the piece that gets overlooked when you put these pieces together is the management and the orchestration piece of that. So I know it's been key to Cisco's architecture from the beginning, but maybe you can give us an update there. Yeah, you know, five years ago when we were talking about the initial integrated infrastructures and that was, you know, V block and FlexPod, you know, it was enough to be able to spin up a VM very quickly because that's really at the time what customers were looking for. They wanted to see, give me a VM, I'll put an application on it and I'm done. We're seeing more and more that customers are now trying to build a private cloud type environment where they want to say, okay, I want to offer you a VM as a service, a database as a service, you know, an instance of an application as a service. And to do this, right, if I can go to a public cloud and get it in 15, 20 minutes, well, I'm going to have to be able to do something like that internally. Now, even with the CVDs, if I have to configure each of the individual components, we can do it quickly, but still it's going to take an element of time. If I can automate all of that, if I can automate the UCS piece, bring it together with the Nexus piece, bring it together with our partner storage piece, you can then get that service up and running much, much, much, much, much faster and you can meet the expectations of the cloud. So the orchestration piece, you know, which in Cisco is Cisco one UCS director and above that Cisco one enterprise cloud suite. We have APIs that connected to our partner storage ecosystem. So now when you say, you know, click, I want to add a service, you can now automate the provisioning of that and it can all come up and running very, very quickly. So with the foundation of it, being the standardized infrastructure of standardized storage for our partners, standardized UCS, standardized Nexus, and then the services that you can deploy on top of that with, you know, never really called it software-defined computing, that's kind of what it is with UCS and then software-defined networking and then what our partners are delivering with storage, it's really easy now to automate these capabilities. Yeah, and you bring up software-defined networking. So the newer component from Cisco is the application-centric infrastructure or ACI. I have to tell you, the whole SDN discussion has been a bit fuzzy. Most customers really haven't gotten it. You know, Cisco and VMware have kind of led in the marketing of it and starting to see customer adoption. It was here at VMworld a month or so ago and it really seemed like security was, you know, the tip of the spear in what they were doing. So, you know, ACI, it's many things, but, you know, what's the key value that customers are having, where's the adoption and how does it fit into this discussion that we're having? Well, at a high level, I mentioned earlier how UCS, you can figure what the server looks like in software and you apply it to the underlying resources. What SDN is doing, or what ACI is doing in the context of SDN is you basically define as I'm leaving my server and I'm going out into the big bad world out there, what does that application require? What does it require for load balancing? What does it require for security as you asked him a second ago? And then how do I automate that? So, if I can say, for instance, here is the web front end to an Oracle eBusiness Suite application. Who can talk to it? Who can't talk to it? How can the web tier talk to the application tier, talk to the database tier? If I can pre-configure that in terms of when you leave the server and you're going to another server, what the network must do there, I can pre-program that with application-centric infrastructure. I can create that template. I can define what that application requires and then I simply apply that under the underlying switches. Now, with the ecosystem that we have for ACI, we have a very broad spectrum of partners around the security realm. So clearly Cisco sells security. We sell firewalls, intrusion protection. We also have partnerships with the Citrix, F5, Checkpoint, et cetera so that we can extend those policies to those partners as well. So therefore, as you're leaving the server going out into the network, you're guaranteed that security will follow that at every point in the network and that's huge from a security standpoint. Yeah, so if we look at kind of the demands on the CIO and IT in general, biggest challenge they have is really operationally. You know, last, God, it feels 10, 15 years we've been talking about up. Spends 70, 80% of our budget on keeping the lights on. Server virtualization unfortunately didn't really move the meat needle for us much. Do you have any metrics as to, how these solutions are helping customers from an operational standpoint? How does it really simplify kind of on day zero as well as during the lifetime time of the product? So let me give you a couple of numbers that we found in talking to our many customers over the years of leveraging and delivering integrated infrastructures. You know, the first is 80%. 80% is about the reduction time that we've seen in deploying infrastructure. So, you know, where it used to take weeks, it can now take days and that's huge. So if I have a team of people or even one person deploying, you know, a flex pod or a V block and they can get that infrastructure up and running within a day, that's a huge time savings compared to where they've been before. Another number I'll give you is 77%. And that's the reduction that we've seen in cabling and infrastructure costs, right? So people often don't realize that the lifespan cost of a cable is about $1,000. And that's the cable itself, that's terminating the cable, that's cable trays, that's maintenance, that's shooting the cable. So I can reduce that cost using the unified fabric capabilities, you know, that are inherent in UCS. That's a huge cost benefit, both CapEx and OpEx. And the other number is 61%. And that's the reduction that we've seen in ongoing operational and management costs for the system. So if I can reduce that, you know, just sitting in front of an operations terminal, right, maintaining the system, troubleshooting and doing break fix. If I can save some of that, then I can put those, you know, very high value employees, I can put them on new projects, right? The first, you know, iPad project, you know, replacing laptops, putting iPads everywhere. Mobility, any of these really kind of cutting edge type solutions, I can now put my resources on that as opposed to the proverbial, as you said, keeping the lights on. Yeah, I'm curious. If you talk about kind of the overall management task there, there's so many repetitive actions that, you know, we don't want to have to deal with. And many of them are just configurations, you know, on the networking side, it's always like, all right, how many hours do I spend configuring VLANs and figuring all this? You guys tie it in with things like ChefPuppet and SmallSaltStack to be able to really automate some of these processes? Yeah, we've been working with all of those vendors. And the good thing there is, you know, you can, again, define what you're doing once and then you can apply that, you know, through the infrastructures you need. And so when you look at the repetitive management task and simplifying all of that, this is really how you get down to the management cost savings and a lot of the TCO benefits that you're seeing with the system overall. So yeah, huge advantage. Switch it over to something policy-based. Make sure I've got APIs to tie in everything I'm doing and, you know, get out of what we was called the undifferentiated heavy lifting, which, you know, unfortunately, you know, keeps folks running around the data center too much and not getting enough work done. Exactly, and that's the whole policy mechanism that we've used, which underlies, you know, the integrated infrastructures. It defines what we're doing with UCS, with Nexus, you know, what we're doing with UCS Director at Cisco One, so that you can deploy these capabilities now much, much faster. All right, so Marcus, you know, if you're talking to your customers, any good stories you can talk is to, you know, how they're really transforming their business, you know, through, you know, leveraging these dev solutions. One of the interesting things that we heard, you know, how it was about five years ago now when VBlock was first coming out and, you know, customers were saying at the time, there's one customer that I remember saying, we love the technology, we just can't consume it. We're not organized effectively. We don't have those capabilities. Technology, awesome, but it's not everything. What we've seen with some of these same customers, some of these very large enterprise customers is that over the last several years they've been able to transform their organizations, which, you know, we talk about technology, but the people in process is really important here too. And a lot of these teams are now really coming together to be able to leverage this converged infrastructure and integrated infrastructure technologies. And that's where they're seeing a lot of the benefits, right, they have more effective teams and they've been able to get applications up and running, you know, within weeks, because the infrastructure itself can be deployed within days. So they can spend, you know, the hard stuff getting the application on the system, up and running, ready to go and out to their users. It is the driver more of the stick because they're competing with the card swipe at Amazon or is it more of the carrot and really seeing the opportunity to free up those resources and do other things? It's actually a little bit of both. We're actually seeing a little bit of both. We see customers that are saying, you know what, I really would like to get control or I absolutely positively have to get control. How do I do that? But there's a lot of forward thinking CIOs who are looking at this and saying, you know what, there's some really interesting opportunities I have. I can really get into this new DevOps type of environment. I can really now work with my application teams to see how can I create the next level of infrastructure to allow them to be more productive. All right, so we're at Oracle Open World. Anything specifically, you're looking forward to this week, presentations, any announcements we should know about? Well, I'm presenting actually in about 10 minutes at the Cisco Theater over here. Always looking forward to- What's going on at the Cisco Theater for the folks that are having to stop by here? So I'm going to be talking about integrated infrastructures again. So for those who just caught part of this, but we're like a little bit more detailed, so they can come by the theater and you know, happy to talk to it, talk about it again. All right, well Marcus, thanks for stopping by, spending a few minutes with us in theCUBE. We are live at Oracle Open World. It's our sixth year we've been here. We're pretty excited to be here once again. We're down at the exhibit floor. Stop by booth 801. You're watching theCUBE. We're going wall to wall for three days. We'll be back with our next guest this short break. Show the value of integration with theCUBE. All right, we're out. That's good. Is that good? I think it's good.