 Live from Gillette Stadium in Foxboro, Massachusetts, it's theCUBE at the VTUG Winter Warmer 2015. Oh, here is your host, Stu Miniman. Welcome back to theCUBE. I'm Stu Miniman with Wikibon, here with SiliconANGLE Media's live production from VTUG Winter Warmer 2015. It's a foggy day here at Gillette Stadium. The Patriots have actually been doing some practicing out on the field, which we're not allowed to talk about or even look at, but the Patriots will be in the AFC Championship this week. Most of the people at this event or locals love to come check out the event, learn from the vendors. My guest for this segment is Joe Anasic, who's the principal engineer from Cisco. Flew across the country to come talk to the over 1,000 users that has spent. Joe, it's hard for me to believe that this is your first time on theCUBE, but thanks so much for joining us. Thanks for having me. I'm glad to be on. Yeah, so Joe, you joined, please correct me if I'm wrong, you joined the NCME Group, before it got the spin in that created the Cisco application-centric infrastructure, or ACI product, and you're now part of the Cisco Proper Group. Give our audience just a little bit of background. What do you do, who you are, and what's your role inside Cisco? Excellent, so I've been with the team, which is now Cisco's INSBU or business unit. My role there is a principal engineer on the technical marketing team. I help work on taking customer requirements, driving them into the product, and then taking the product and driving it out to our customer base. Cool, so Joe, I also come from the networking world, and one of the challenges I usually have at an event like this, or if I go to VMworld, is it's a very different audience that you'd find at Cisco Live. Cisco Live, these are guys that, they know how to cable stuff up, they know how to configure things. When you start talking about IPv4, IPv6, and all the other protocols and terminologies out there, they live it day to day. You go to a virtualization type environments, and the networking is that thing that the network guy won't fix the quality of service for, so it's something that is nothing but a pain point to them. So this whole kind of ACISDN thing, why should anyone outside of networking care about it? I would look at it from a couple of different perspectives, but if you think about the deployment lifecycle of a new application, or development lifecycle of a new application, the network today, including the Layer 4 through 7, to deliver that application securely and with load balancing user experience, is the slowest piece to change. You can spin up virtual machines in a few minutes, a few seconds, storage pooling, maintains very quick storage provisioning, and then you get down to IP subnets and VLANs and all these other things that we talk about on the network side. Yeah, so just a quick point, your Layer 4 through 7, it's things like your firewalls, your load balancers, all of those applications in the network, right? Absolutely, yeah. So I think, I look at ACI always jokingly say, if we did our job right, we make the network disappear. We allow you to provision the applications and services you need to at the time you need to do them, and you'll never notice that the network's there. Yeah, it's similar, you know, there's the analogies and I don't want to get into the terminology bait about, you know, network virtualization and SDN and ACI and everything, but server virtualization made me, for the most part, not think about my server anymore. And what we want to do is not have to think about at least the switch as a single device anymore, really, I think in your session I liked, you talked about it. You know, I don't care about the phone, I care about, you know, it's the SIM card and the data and everything I can do with the phone, it's not about a device. And the same thing is managing, I don't want to manage a switch, I want to manage networking, is that a fair statement? Absolutely, and it doesn't matter if you go with ACI or one of the competitor's products, the idea is you want to be able to look at your data center network as a system for application delivery and provision it in that type of fashion. All right, so Joe, where are we? We've been talking about, you know, this revolution in networking for a long time, networking tends to be one of the slowest moving groups in IT. I mean, we're still talking about 10 gig adoption, you know, a standard that was ratified in 2002 and still today, a little bit more than a third of all servers are going out with 10 gig, it's much more prevalent in the core, but you know, I usually measure change in the network in decades. So, you know, where are we with the networking revolution? I think we're still pretty slow in where it's being adopted. If you look at, a lot of the problems with the network is that we're so slow to change. For 20 years, we've been using IP and spanning tree to fix every problem and, you know, it's like I was talking about earlier in the session. I think we've got a lot of adopters. There are people starting to get their feet wet. There are a lot of people getting into the proof of concept stage and a lot of people starting in small production environments, but we're still early stages where, you know, where cloud was maybe eight years ago. Wow, eight years ago, you're going back to, you know, it was Amazon and a couple others, but not too much. Is this discussion mostly at the network level, is it getting up to the C-suite? You know, what's the core value business proposition that you get to companies that you're talking to? So, the conversation I like to have is going to be from the C-suite down to the application team or developer teams, because the idea is that we really want to accelerate their life and how they're able to deliver business demands. One of the things I love about the industry today is that IT's really coming back to a point where we're a business differentiator. We're not just keeping the lights on. We're not just the power company anymore. We have the ability to change our business, to change our customer and users' lives. And so, if you're talking to those teams about how you can accelerate their deployment life cycle or bringing on new services, I think that's the right level. Wow, it's real interesting, you know. Applications are the reason why we have infrastructure, yet it's not traditionally something that the infrastructure people have been involved in. What does this mean to the life of the network admin, you know, the CCIEs that are out there? What's going to be different for their world? So, from their world, I think the jobs aren't going away. We're still going to need those people that can do that. Even with an SDN solution, you're going to have to get down to the weeds and do some troubleshooting and know the routing protocols when things aren't working correctly. I think they're going to have to learn some new skills. They're probably going to want to have some understanding of how APIs drive a product. They're going to want to have some understanding of basic scripting, but this is stuff that they probably have some exposure to in the past. I think just like anything else in IT, we have to continue to adapt, otherwise we become irrelevant. Yeah, that's a great point, Joe. Boy, yeah, I'm just trying to think. It's, you know, talk a little bit about, you know, your day-to-day role, you work with the channel, you talk to customers, you know, what's a day in the life of Joe? A day in the life of Joe, usually I'm trying to get myself fired with what I tweet. In between that, I'm talking to customers. I spend a lot of time putting together collateral and documentation, things that we'll go out and present, white papers and that. And then I work very heavily with our Cisco channel partners. Somewhere around 80% of Cisco's business goes through our channel, which allows us to scale and get much closer touch with our customer base. And that's where I like to focus my time. Yeah, so, you know, we talked about, it takes a long time for things to develop and people are, you know, adverse to change. But one of the things we have, we need to get people out of their comfort zones. And once they've adopted it, usually if the solution does what it's supposed to, my life should be better. Can you paint a little picture for us? You know, what's a customer say when they're done that they're like, oh, if I'd only realized this, I wouldn't have fought you for six months on trying to make this change. What's kind of the aha moment? Or you know, the real takeaway that transforms businesses after they've done this? So I think from an ACI perspective at least, one of the things that gets that aha moment is that idea of your business has requirements for the applications and services you're deploying. We take that and we translate it automatically into infrastructure provisioning. So you get rid of the translation layers that happen from human to human. You're reducing the risk of error as that gets deployed and you're able to tie that information back up in a relevant way, whether it's through our health scores or our troubleshooting telemetry or those things. And I think when they get that idea, if I can take the business requirements of SAP HANA for instance, and use an application network profile to deploy those down onto my network, meeting their best practices in an automated fashion, it clicks. Yeah, it's interesting. I think for, you know, for so long we've been hearing IT say, you know, I just need to do more with less. I've got no budget. I can't do anything. But, you know, I'm spending, you know, so much of my time just, you know, banging my head against my infrastructure that I have and it's not my friend. Does the picture look better in the future? I think it does. And I think we have to make it look better. It's not just that we have to be able to continue to accelerate what we put onto the network from an application and service perspective for our own industries and against our own competitors, but also to keep our own users. I mean, the whole idea of shadow IT, if I can't give you something you need to do your job fast enough, you'll swipe your credit card and find it. And then I have no governance control, security or risk compliance of that. All right, I heard you, you made a comment at the beginning of your segment that, you know, the network people in IT, they're like, you know, some call them ninjas, I think you call them MacGyver. Is that you kind of throw a bunch of stuff at them and they're going to get the job done. You know, the network is really, their whole job is just keeping everything running. And, you know, it's, if you saved a little bit of money, that might be great. But if I go down, you know, you're out of a job. So if the MacGyver was the old world, you know, what's the new world? If yeah, I like that. If you go down, you're out of a job. I always say the best disaster recovery plan for a network engineer is a fresh resume. But what's the new world? I think that the new world is a network that's looked at as a system, not a bunch of boxes. Regardless of the vendor you're talking to, a network that's there and can deploy what you need it to, when you need it to, and then give you back the visibility and information you need. Things are going to go wrong. Networks are going to go down. Ports are going to flap. Things like that happen. But we need to be able to identify where it's happening, remediate it, keep things up and meet our SLAs. All right, how much of this ACI discussion leads into the kind of cloud and mobile world? I have a lot of friends in the kind of the wired, you know, layer two, three, you know, switching world. And they said, you know, I'm never touching a switch anymore. It's all Wi-Fi is what they've gotten there. And Cisco's got its inner cloud. How does the cloud mobile world tie into what we're talking about in the data center? Yeah, so I think the ACI and inner cloud story actually work quite well together and that's a lot of what we're working on right now. If you take a look at the idea of the inner cloud from Cisco is giving you the ability to tie your two data centers together, whether it be private clouds or public clouds or both, what you need from there is the ability to have consistent policy as you move applications back and forth. So because we're packaging the policy is that logical entity within an application and profile, we can then migrate that across to your public environment, to your other private clouds and make sure that when an application is running in certain areas, it's got the policy requirements and security requirements it had originally. All right, how about really kind of the management and orchestration of all this? Network is a piece of what we do. We used to have the joke that, you know, single pane of glass is spelled, you know, P-A-I-N because it never works that way, but, you know, how does networking, you know, is networking still going to stay separate or is there kind of management and orchestration across the board that's going to manage this for me? So you're saying that the network isn't the center of the universe? There's other things? There are other things. All right, so for the other things, there's a lot of different options to go with and I don't think, you don't want your network tool to be the one that's running all of your things. So systems like OpenStack, like CloudStack, great ways to plug in and tie in and manage that network overall. There's a lot of other tools out there. Cisco has several flavored tools and a lot of the other, you know, VMware sitting here in a virtualization environment, they have a lot of great tools in that space as well. What you want is the underlying infrastructure components to have the most programmatic interface they can so that they can plug into that next device and you can start to pull all your resources together. All right, last thing I want to poke on, Joe, is you talked about open source in your keynote this morning. I want you to talk a little bit about it and how can I truly be open if I require you to use my hardware? That's an excellent question. So my hardware-defined networking? Yeah. So if you take a look at what we built, we built three major things. We built our policy model, which is our way to describe applications. We built a controller that's able to deliver that and then we built the Nexus 9000 hardware, which you don't need all Nexus 9000 to go run ACI, but it is a subset requirement. If you take that policy model, or description of applications, we've open sourced that model and worked with the Open Daylight project to have that in the group policy objects there. So you can take the same way we describe applications, use that with the Open Daylight controller, throw that in with OpenStack Neutron over the OVS and not send Cisco a dollar while still applying the ACI concepts with the way in which you develop. We want to make sure, I think from my perspective, that if you go down the path of ACI today and you start to rethink the way you deploy and design applications, that you're not stuck buying Cisco no matter how much we change over the next five years. That you can take that same application mapping, take it over to an open source platform and keep moving. Yeah, talk to a lot of users out there. There's the Open Networking User Group, one of the key things that some of the big banks and large enterprise customers were saying is what they'd love to have is really an enterprise application store for most of their environments. Software to find storage in some ways is going there. If I can separate all of my services from the hardware and be able to choose pieces of it. Does Cisco agree with that vision and are we going to be able to allow customers to just buy network services from a lot of different people? I think Cisco agrees with that vision. I think it's a big company with a lot of market share and networking probably isn't going to be the fastest thing to go down that path. And I can't speak for the company as a whole, obviously my little legal disclaimer. But if you look at even the Nexus 9000 platform, it's already an open hardware platform. We have several operating systems you can run on that platform. We could definitely take that, we potentially take that the next step. We could allow another operating system from a third party to be run on it or allow the operating system to go out on its own. Or even, I would love the idea of a networking operating system that I download like an app store. I need BGP and OSPF, so let me click those and that's the only code that gets put in. But I think I got a pipe dream on that one. All right. Well, hey Joe, last question I have. Why aren't I seeing you writing more? You used to write some really good stuff. As a matter of fact, if I remember right, you won an award from Wikibon for being a positive deviant or what we called a pause dev contributor to the world out there. So, you know, is the job just keeping you too busy? So the job kept me quite busy for quite a while. They're starting to offload a lot of the administrative overhead of my role which gives me some more time to do that stuff. And then I think I need to make sure that as I continue to write more that I'm a little less of a vendor stooge and a little more of the guy that got that pause dev award originally. Yeah, absolutely. Joe, you always appreciate you've got some great insight. You know, you can wear the corporate badge some because that too pays the paycheck. But appreciate you coming, talking about the real issues, even brought up some of the terms that I didn't have to on here. So have you back anytime. Thanks for joining so much. We will be right back with lots more here from the VTUG right after this break.