Big Switch Networks CTO Rob Sherwood on SDN in 2015: The Time is now

Big Switch Networks CTO Rob Sherwood touts the maturity and ease-of-use of Big Switch SDN solutions, the power of open switching platforms, docker, overlays and more

1 2 3 Page 3
Page 3 of 3

[Art] As open networking takes hold, it looks to me like innovation in servers, innovation in networking, these manufacturers are really wrapping their hands around them now. It's not just, I am a networking manufacturer. I am a server manufacturer. I tend to think if I buy a new server and network combination today to try and get my most efficient private cloud combination that I can; three or four years from now I’m probably just not going to upgrade the server. Due to the pace of that network innovation that's happening, we're doing a 100 gig and soon can do 400 gig and not to mention the other feature capabilities. I tend to think that we're going to see a lot more of this type of architecture where this year I'm going to deploy this private cloud and maybe a few years from now, or as we get into the next paradigm we'll do the same thing again. We'll stack up another new one. Do you see that as an architectural trend?

[Rob] Probably so. So much so that we started running with the term we heard from our hyper center friends, which use a core and pod design. The idea that you have a routed core which is your ingress and egress to the data center, that lives through the lifetime of the data center. That might be there for ten years or whatever that number is. Then you have a lot of pods hanging off of it. A pod is, think of it almost like a row, but it's a collection of compute, storage and networking gear that's all certified to work together. The biggest selling point of the architecture is you've got a team whose job it is to constantly be working on the new pod design. They'll have the pod design v5 and while v5 is the latest greatest, every time you need a new project, you’ll stamp out another pod of v5.

Then while the team is working on it, they'll figure out, "Okay, what is the next best version of compute, storage and networking that ties this all together? What is the next design?" Then when v6 comes in, they'll stamp that out every time they need a new project. This is kind of multi-use, multi-purpose architecture so it can have lots and lots of workloads in a pod. The real value is actually in the automation. If you were willy-nilly picking up, every time there's a new rack deployed and you've got new servers, and new storage, and new network then the automation of that is horrible. When you actually have some homogeneity, which is to say I know this is pod v4 then that's much, much less complicated to manage. That's where people are really seeing, both, the value from a purchasing cloud standpoint as well as from an automation standpoint. There's now really a huge shift in how people are building data center.

[Art] Let’s say you're an enterprise customer and you're looking at deploying private cloud technology. You want to have things that are automated, and elastic, and what have you. What is your take on physical OpenFlow and physical hardware versus NVO? If I'm a customer, do I use both and they interact in some place? Do I choose one or the other? What would your advice to customers be around that?

[Rob] Well, there's two dimensions which is, one is, what is your workload? Depending on your workload, you can probably virtualize it but you may actually have some physical hardware as well, for example some database servers and things like that. Most of the customers that we talked to have kind of a 70-30 split, which is to say 70% of their workloads are virtualizable and about 30% are physical only for various reasons. Understanding your workload is the first place to start. Most people end up with some sort of mixed workload or can't know ahead of time. The answer is you would probably need both. The other side of it, it's not really an either/or decision. Which is to say, if you have and NVO solution, the packets don't match what they're getting from one vSwitch to another you need a physical network there. If you're going to need a physical network, the question is, even if the policies are maybe not that complicated, having a single point of control across the 40 switches is still easier, and cheaper, and better.

[Art] I am a fan of NVO technology, don't get me wrong, I don't have too much an opinion on it one way or another. Sometimes I think like, would the world really be that much of a different place if Betamax had won instead of VHS at the end of the day? I’m not saying they're the same quality of technology. I was thinking it's probably possible to make either one work if that's what the industry chooses to put its weight behind it. I think what I’m waiting to see is one going to take over from the other. To my point, one thing that seems apparent with OpenFlow in the physical hardware is, yeah, today it hasn't been to full maturity yet and I can't just go out as an average enterprise and go buy a plug and play solution [for all use cases]. That hasn't happened yet but that doesn't mean that OpenFlow technology won't ultimately create that. Once I have OpenFlow technology in my physical hardware, is it that stuff starts to become overhead that you don't need anymore?

[Rob] Let me push back a little of what you said. We've been shipping OpenFlow based solutions that really are plug and play for the better part of two years now, particularly our Big Tap product. That really is a drop it in and it goes on the side of your network very safely. It's a very safe and easy thing to do. We're getting a lot of traction with that. If that builds on OpenFlow, we've been deploying that for a couple years now. We've been deploying our big cloud fabric product since September of last year. We're getting a lot of traction with that. Actually our new release, if I can put in a plug for that, is coming out this month and it's going to actually support hardware from Dell so we're really looking forward to that. For 2015 it's going to be very interesting to see now that we've got some real initial customer traction and we've got some real products that will start to just connect all the dots and show the hockey stick curve. It'll be very interesting to see what happens this year.

[Art] The next thing that I wanted to ask is really about Docker. That's one of the big new fairly uncharted territories. Linux containers have been around for a while, the networking capabilities within Docker aren't particularly robust yet [This call was recorded prior to the Socketplane acquisition by Docker]. There's a lot of startups trying to tackle that and I'm certain it's on your guys’ radar and you're working on stuff around that. I'm curious, what do you see, how is Docker networking going to hit the market and what evolution needs to happen there?

[Rob] What's fascinating to me is, Big Switch has a virtual switch and we’ll be shipping it later this year. The virtual switch literally cannot tell the difference between a VM on one side versus a Docker container. The way that the Linux virtual Ethernet works on that type of thing. From a networking perspective, Docker doesn't really change much. You still need something that looks like a vSwitch, with the functionality that a vSwitch has, but better. It actually increases, if you want to think of it in these terms, it's kind of the VM density so you can actually get much higher density with Docker. You can spin them up a little bit faster, it's a little bit lighter so maybe there's a little bit more churn on that side but a lot of the fundamental policies don’t' really change. This is to say, most people treat a collection of VMs as an application, Docker just lets you do that faster and cheaper. All the policies and the user experience and the management objects and things like that, all of those things will stay the same.

[Art] Well, that's great. It just seems ... I think that's also really, particularly for the enterprises, if you're at the leading edges of web tech, you're going to see a lot more of the warts of the technology, if you will. I think, particularly for the enterprise, it seems to me that things are getting really streamlined for Docker to where pretty quickly the management and orchestration stuff that surrounds VMs today will probably be equally applicable in most systems for Docker containers tomorrow. Does that sound about right?

[Rob] Yeah, absolutely. It's, honestly, I definitely see the benefit of the data center industry converging in terms of storage, compute and networking and all of these things but people are still learning the skill sets. Honestly, I think it's the case that companies are coming in to solve specific problems and are not necessarily aware of the problem from the other side, the facts as it were. Being able to bring the solutions wholesale from the other side of the fence I think will really help accelerate the adoption of these types of technology.

[Art] If you're a business, you need to take part in this internet of things thing, pretty much no matter what product you make you probably fit in that somewhere. You need to be aggressive, you need to be innovative and it seems to me open networking, SDN is really key to that. I'm curious about your thoughts?

[Rob] Absolutely. Honestly, that's actually what gets me up in the morning. That's really what motivates me, which is you look at the explosion of applications that happened when all of the sudden people could write apps for smartphones, or even further back when people were moving from mainframes to PCs where people could invent their own programs. You're actually enabling everybody in the world to become an innovator here and it'll be fascinating to see what happens. At the same time with internet of things, there's kind of unprecedented security issues. The idea of having my toaster or my garbage disposal online is a little bit scary.

[Art] If I could interject something there, on that point about the toaster. At the consumer electronics show this year, now the Roombas instead of just being a vacuum, they're remote controllable and have a camera in it. I don't know if it's the Roomba brand, but most of the ones at the show floor that were similar, the robotic vacuums they have a camera and they're meant to be like home security system integrated now. When you're at the office, I can tap into my Roomba, drive around my house, view everything in my house. It starts to really, that's even still a very, very small example but I think it highlights your point about the security implications. Any hacker can now just not get into our personal information, they can come into our house; that's a problem.

[Rob] Absolutely. Security is kind of a big tent but what it translates to into more specifically what we're talking about is more and more complex network policies; who can access that camera during what time of the day, on what networks, and things like that. I think those have become the ... As those security policies compound, that's really what makes network management so complicated and that's what I think really forces us to network automation. I think that's really the only way to produce solutions to this.

[Art] I think it's a great segue to end it there. The thing about this internet of things, this realization that every business no matter who you are, you're going to be part of this and to win in it and to hopefully win these 50 billion new devices and 5 billion people that are going to be coming on the internet over the next few years. There's a lot of opportunity to win in it. I think open networking, software defined, modern web-scale architectures are the only things out there that really give you that flexibility. Big Switch has I think been the, you've been really the first and the main one that really stuck to the core of OpenFlow. I think you have a great solution. Look up Rob, I'll put the links up here to follow him. Thank you, Rob, really great advice for the audience today. It's been a great conversation.

[Rob] You're absolutely too kind, but thank you so much for having me.

Join the Network World communities on Facebook and LinkedIn to comment on topics that are top of mind.

Copyright © 2015 IDG Communications, Inc.

1 2 3 Page 3
Page 3 of 3
SD-WAN buyers guide: Key questions to ask vendors (and yourself)