 Wikibon headquarters, and we're talking about software-defined networking and what we call the Software-Led Data Center. VMware's acquisition of NYSERA for a billion dollars really was a catalyst for changing the discussion around networking. There's a lot of forces at play. You've got flash eliminating spinning disk, which has traditionally been the bottleneck of the system. And now that bottleneck is getting pushed to the network. Networking for the last four, five, six years has kind of been a boring space. Cisco dominates. You've got some others like Juniper and Arista nipping at their heels. But all of a sudden networking has become the hot sector. And software-defined networking is really the focal point now of value creation. And we believe that there's going to be a ton of action here. And I'm here with Stu Miniman, lead analyst at Wikibon, principal contributor. And we're going to talk about this trend. Stu, welcome. Thanks, Dave. So we saw the big switch announcement the other day and that added fuel to the fire. Take us through what's going on here, Stu. Sure, Dave. So as you mentioned, networking's been kind of boring. But if you look at the market, there's actually been a lot of money to be made. Now, of course, Cisco is making the most money in this space. Not only are they dominant in market share, but they're also making huge margins. And if you look, people said it's kind of proprietary hardware, big iron. And the big trend that we're watching is, as you said, Wikibon and SiliconANGLE looking at software-led infrastructure. So how can software really decouple from the sheet metal, the physical environment, to really make scalable, more flexible environments that are really built for kind of virtual and cloud environments? And that's kind of some of the promise of kind of this umbrella term of software-defined networks. Now, big switch networks is interesting. On Twitter, the discussion lately has been that big switch networks is probably one of the worst stealth companies that's been out there just because most of us have been talking them for the last year or so. Wikibon teams talk to them quite a bit. You've got some great founders in the big switch networks group and really some guys that six months or 12 months ago were really super glued to open flow, which we'll talk a little bit about. And really have pivoted quite nicely to have this blanket launch, which really goes into their vision for software-defined networks and how they're going to transform networking. And as I said, it's all software. There's no hardware. A big switch network doesn't make a single switch. And what is that impact going to be on the switching infrastructure itself? So essentially, if you're putting forth a vision where the hardware is standardized, some people call commodity hardware, and you're basically software is controlling the flow of data, right? That's kind of the software-led vision. Yeah, so if we look, the commoditization of IT has been one of those big trends. Everybody gets to the same type of device. Everybody sells the same kind of disk. And if you look at networking, one of the big pieces that are what we call the merchant silicon. So Broadcom's leader in this space, Intel picked up Fulcrum Micro last year. And Marvell are the ones that make the standard chips that most of the guys other than Cisco and Melanox and Brocade are making their own chips. But everybody else uses the same chipsets. So this is sort of an attack on the custom ASIC proprietary chipsets? Yeah, absolutely. And what's interesting about big switch networks is one of the things I'm trying to parse is that big switch networks, SDN and the overlay networks, they don't change necessarily how many ports I use. And it does allow customers to choose what they want. But maybe those multi-vendor environments can grow a little bit. And it's that battle for where does the value live? Is it the vendor that makes their own chips and bundles everything together? Is it Cisco or is it somebody that takes merchant silicon with their own software and then partnering with somebody like big switch? So that whole, we always look at that kind of pendulum swinging between one vendor that puts everything together versus pulling apart the pieces and who adds the software that adds value to this. So what does this mean for a company like Cisco? I mean it's no secret that Cisco was looking at NYSERA as well, VMware ends up with it. What does it mean for a company that has for years been able to create sustainable competitive advantage by developing proprietary capabilities and monetizing that value? What's the future hold for guys like that? So first of all, if SDN is the future, Cisco's behind. So Cisco a couple of months ago at Cisco Live put forth their whole vision for what they call their SDN strategy. So if you look at what makes up an SDN solution, one of the pieces is an open flow controller. So this is something that we knew big switch was coming out with and that they launched with this piece of the technology. But big switch isn't the only one to have an open flow controller. VMware with the NYSERA acquisition as an open flow controller. NEC started shipping a controller and there's already announcements out from Cisco from IBM and HP that that controller which sits between the physical switch or virtual switch and the application. So this is where kind of the upper stack that's going to control all of those messy networking protocols and allow things to work in hopefully a more automated and simplified manner are going to be controlled. So if we go into the announcement that big switch networks had, it's the big switch controller, which the kind of exciting piece that they have there is it's built with the same API set fully API compatible with the Apache project known as floodlight. So floodlight is something that it was mostly the guys from big switch that helped create that technology, but anybody can take it. So there's been over 10,000 downloads out there, and then big switch kind of makes their own version of that packaged up. So if I did all my testing and started building my applications on floodlight, I can go to production using big switches controller. How does big switch compare with say NYSERA? Okay, so NYSERA has what they call their MVP their network virtualization platform and NYSERA can use open flow as part of what they're doing. It's a tool and that's I think one of the kind of refreshing things is what I really liked about NYSERA and where big switch is pivoted. Open flow is something that helps with all those kind of messy protocol transitions so that I can take kind of the physical and virtual interactions and all these kind of layer two layer three overlay things like NVGRE and VXLAN and should help pull the control plane away from the data plane. So really simplify that network for the administrator out here, but open flow is just something that lives down at that protocol level. If you're not a networking geek, you're probably not going to care that much about it. It's that broader network virtualization, how we abstract the environment and manage that whole environment, which is what big switches piece of that is, let me see, it's the called the big virtual switch. And that's really the answer to NYSERA's MVP. So I'd say there's a little bit of overlap between NYSERA and big switch, but it's not like they're completely mutually exclusive. It's definitely going to be that war for who gets the footprint and who's going to control it. And the third piece, of course, is management, which is kind of the holy grail and all the network management and big switches third piece is what they call big tap. And that's where there's big money and there's big control over the environment. So, you know, VMware with the bee center that they've got out there and how NYSERA is going to help with the management and what big switch now has with big tap. You were talking earlier about big switches APIs that they announced. So that begs the question about the ecosystem. I think they made a big deal about the ecosystem. Can you talk about that? And what do you think the impact of that is? How real is it? Give us some color there. Okay. So, yeah, excellent question, Dave. If we look at kind of the networking ecosystem out there, of course, there's Cisco dominating and then there's a bunch of other Ethernet players that have been getting into niche environments or places where they have a competitive advantage. So, big switch has come with a good, you know, group of ecosystem partners. So, Arista, Dell Force 10, Brocade and Juniper all have formal partnerships. And of course, these are the guys that today have not announced that they are going to do their own controller. They also have interoperability with IBM and HP who have said that they're going to have their own controller. But as we know very well, IBM and HP are very good at playing across the board. They're not just going to do VMware. They're going to do Hyper-V. They're going to do Zen. And from a partnership standpoint, two other pieces that I thought were critical from the big switch announced. One is they've got formal relationships with both Microsoft and Citrix. They will work in VMware environments, of course. Things like VMware's V-switch and the Open V-switch. Big switch networks is going to work with those because if they don't work with those, they're going to be missing a huge part of the market. But the second piece is kind of the upper layer of the stack. If you look, big switch network is working closely with the OpenStack ecosystem out there, people like Canonical. And on the layer four through seven, they've got folks like F5, Infoblox, and even storage guys like Corade that they're partnering with to kind of have that full application integration. Is Riverbed in that list? I'm looking at the slide and I don't see them, but I thought somebody had mentioned Riverbed. Yeah, we think they would play there, right? Yeah. Who is actually buying stuff today? Is there any action in the market? There is multi-billion dollar question, Dave. So if we look at NICERA acquisition, they had some of the big telcos and big financials, guys like Fidelity and NTT and Telstra. So when I looked at kind of the logo deck that big switch networks came with, of course these guys came from Stanford and they've got a lot of universities and they've got all of these development shops that are playing with Linux and doing this. But it's moving from kind of the universities who a lot of them are playing with this stuff and even have a little bit of government money to play with this stuff into, you know, real enterprise environment. So I have heard some of the, you know, financial guys, I believe I read an article that said Goldman was one of the ones that are playing with this. And I've been reading some of the articles that say some of the customers, but big switch hasn't yet rolled out, you know, here's all of our customers. But the ones that are doing it are big customers that are buying new equipment or can build a kind of, you know, a patch of green where they've got switches that support open flow because here's one of the things. I can't just take, you know, my three-year-old, you know, Cisco, you know, catalyst switch necessarily and work in this environment. I have to switch and, you know, end devices and, you know, virtual switches that all support open flow. So this is going to take time for our typical enterprise customer to start looking at this. So we talk about a lot, we keep on talking about GRS, getting rid of stuff. Will this whole SDN trend allow customers to get rid of stuff, and if so, what? So it's getting rid of stuff by adding. So that whole, you know, how do we, you know, I've got this switch in this environment with these kind of, you know, protocols, you know, how do I get beyond that? So it's this layer of abstraction or indirection that should help simplify the environments. Now, of course, I was actually going back and reading an article that Chris Hoff wrote a couple of months ago. So, you know, Chris Hoff, good friend of the Cube, beaker on Twitter, and he said, you know, if a customer, enterprise customer just goes to the cloud and if the cloud can host my environments, I don't need to worry about this. So this is kind of a middle step. How do I look more like, you know, the Googles of the world and manage my environment? But it's not a, it's not like all of those networking problems go away. And the thing that I'm trying to poke out a little bit is what we really want to be able to do is get out of our silos. And when we talk about what server virtualization with VMware did is, you know, really created that virtualization administrator to help manage more of the data center. And we want customers to be able to not worry about managing LUNs or, you know, managing whether I'm doing, you know, how many VLANs I have and how large my layer two is versus layer three and all that stuff. I want to have data center administrators. I've been talking to some interesting startups out there in the Valley that are looking to transform networking even further. Talking about guys that are using the ODM switches and truly building off of, you know, the merchant silicon with new versions of software. Because if I look at what Cisco's done and all the guys that are attacking them is, you know, they've all got that feature creep. It's the innovators limit. They're moving up the stack and have the really big customers and keep adding features. I mean, just pages and pages of how many protocols they have and how they build their environments. Now, there's customers that need the solution, but we've really over architected a lot of what's going on. And for many customers, we could really simplify the environment and that stands part of the answer, hopefully. And that's the right direction to move towards a software defined world. But, you know, if a customer rather than saying I want to redesign my data center, I want to move many more of my applications off site. You know, that could be simpler. What about the cloud service providers? Are they all over this? So we were at Brocade a couple of months ago. And, you know, you did hear that, you know, Brocade's one of the partners of Big Switch Networks. And their service providers are starting to look at, you know, these kind of technologies. And Brocade's definitely looking at service providers for SDN. So this is part of what they're doing. You know, when we did our recent peer insight with, it was Luke Norris from, help me out, Dave. I don't know, but anyway. Yeah. So, you know, the service provider to service providers. So, you know, they are, you know, just such hyper growth. You know, they've spun their network a couple of times and are always looking for ways to simplify it. So, Pete Collo. Yeah, thank you, Pete Collo. Sorry, Luke. Okay, so my last question to you is, you're a CIO. You're hearing all this stuff on SDN. You see the NYSERA acquisition. VMware's making a lot of noise about this. Big Switch making a lot of noise. What do you advise the CIO, what should they do? Should they just wait, put it all in the cloud? Should they hop on this trend? What's your advice? So I think it matches a lot of what we've been telling CIOs in general is, you know, we need to start, you know, educating our workforce for the next generation. If, you know, you've got guys today whose jobs on storage were provisioning luns, you know, their jobs are going to change within the next couple of years. Same things on networks if I'm doing, you know, my zoning and my VLANs and, you know, really down in the weeds on some of this, we need to be doing some cross training. So understanding the applications and what they need and how do we support them getting closer to the development side of things, getting closer to the applications. You know, my advice for CIOs when I wrote about SDN a couple of months ago is like, look, you know, most of you are not going to look to deploy this this year. As a matter of fact, in 2013, you're probably not going to look to deploy this, but you might be going to service providers that are doing it. And, you know, it's something that over the next 18 to 24 months is going to be kind of pushing down and, you know, will become real as the ecosystem matures. So, you know, learning, reading, training, you know, getting your folks not to just, you know, cling to the old ways of doing things. We need to learn how to move forward on these new technologies. Excellent. All right, Stu. Well, thank you for taking us through some of these trends. This video is going to be up on SiliconANGLE.tv. It'll probably be up on YouTube as well. Stu's going to write this up if he hasn't already. SiliconANGLE's going to get this covered like a blanket. I know John Furrier's got a piece coming out on this whole trend. There's a slide share that went up today that I saw. So check out the SiliconANGLE site on Slide Share. And we'll be covering software-led infrastructure. It's a major focus of us here at Wikimon. So thanks for watching, everybody. Stu, thank you for coming on. Thanks, Dave. And we'll be back next time. This is Dave Vellante live from Wikibon Headquarters in beautiful Malboro, Massachusetts. Thanks for watching.