 Live from Aluelto, California, it's The Cube at Pier 2.0. Brought to you by the Pier 2.0 Foundations. Learn, connect, and grow. Now here are your hosts, John Furrier and Jeff Frick. Hey, welcome back, everyone. We're here live in Silicon Valley for The Cube. This is our flagship program. We go out to events, extract the seeds and the noise. I'm John Furrier, the founder of SiliconANG, joining me. Jeff Frick and we are at the Pier 2.0 event, day two. Wrapping up here on an amazing event, inaugural event for the group. Huge success, not huge numbers in terms of audience, but very small kernel of experts and gurus, really setting the agenda going forward. Our next guest is Keith Mitchell, president of DNS, OARC, the operations side underneath the DNS, global viewpoint, welcome to The Cube. Thank you, John. You know, The Cube, we always love the global perspective and obviously the DNS is running the internet. It's a lot of traffic controls through DNS. Operations is a bit interesting term. The word dev ops, we've been covering on the cloud. It takes a more development focus around operations. But developers, they make a lot of errors, agile programming, ops is not something like the tolerate errors. So some people say ops dev. So operations and developers working together seems to be an oil and water situation. How do you view the state of the current communities coming together because you have the two collisions of the two worlds, the developers, software developers and the ops guys. Is there harmony there? Well, that's an interesting question because when I started my career with software engineering, you know, and I was writing TCPIP code 20 plus years ago and I discovered very quickly that I am a terrible programmer but I'm actually pretty good at network operations. So I would be kind of wary of somebody who said they were good at both. There are distinct skill sets there. So I've done my time in network operations and now what I'm trying to do is help facilitate education, transfer of knowledge, distribution of clue within the internet operations community and various forms. Subscribe to that in a little bit more detail. Drill down on that distribution piece. What is it all about and what's the key points? I think that it's about running the infrastructure, finding models for running the infrastructure that work, defining the processes, finding the talent and facilitating the ways and make the talent work and then building the infrastructure, not just in terms of their routers and the switches and the servers but also building the infrastructure in terms of the organizations and making sure that these organizations are functional. They adhere to best practice. They are good for making sure that the internet works effectively. What's the critical success factor for these young guys coming in? Because education, people are starved for education. There are new recipes, new formulas, new network architectures, you've got virtualization, you have large scale networks that are developing fast. What is the key issue that people are most interested in? I think that today's event is about being about peering and peering is like the secret sauce that holds the internet together. There is a bunch of best practices and a bunch of people who have been doing this peering for a long time and we've learned some pretty good things but we're not perfect and we're in a very dynamic and moving landscape. So I think it's very important not just to preserve the knowledge that we have but also make sure that the processes and the structures that we put in place are open to the innovation and the changes that come with social media and the cloud which are things that 20 years ago we had no idea of. We're building something for the ages here. It's not just this week or this year or the past decade. Yeah, and it's interesting when we were talking a little bit before we started, you said that this is the internet secret sauce and that the really the number of people that really know how it works and are making it work and driving it is a relatively small group and that the purpose of Peer2.0 is to start to get the education out to a much broader audience to get the messaging out to a much broader audience. Where did that come from? Did people just wake up one day and say, oh my goodness, you know, there's just not enough of us or have the demands of the network really forced the hand to say we've got to get this out to a broader audience? I think that there's always been a recognition that the knowledge needs to be shared. I think that as we have seen everything evolve over the past few years, we've got to the point where I think that a lot of us are realizing, well, actually, we're in danger of some of this knowledge being lost. We're seeing some of the people who are mentors are starting to retire and we're thinking, well, it's important that this is preserved. And the other side is there's huge demand for it. I've been running a UK network operators forum for 10 years ago, that's for 10 years now, which is an organization that is equivalent of Nanogue in the UK. And the last few meetings we've been growing from 150 to 200 to 250 attendees. And that's just in a small country like the UK. There is huge demand for this understanding of how do we make all this hang together? What is the glue that underlies all the social media and the cloud services, the critical infrastructure, making sure that that's available and functional? And is that a function of, we talked a little bit earlier about, people are now peer-to-peer directly. The ecosystem of the peering and of that core network, internet is expanding beyond what it was before just the tier one guys all peering together. Is it because you're seeing more interest from enterprises or companies that want this capability or is it just pure growth within the infrastructure that's driving this demand? I think that there is, there's not new players or traffic so much as everybody is realizing that more and more of everything that we do on particular businesses is reliant on the internet. And it's about ensuring that the connections are more robust and resilient and controllable. I mean, one of the key things about peering is control over your own destiny. So that if you've got another business that you were doing business to business interaction with, and it's critical to both your businesses that that traffic flows, and you do not want the cable company in the middle to start getting involved in network neutrality wars with your traffic or you're worried about a backhoe going through the single piece of fiber that they may designate for your traffic. So by having your organization peer directly with the businesses that you have a critical relationship with, you are having more control over your destiny. And that's exactly the way it's always been. Internet providers started peering not just because they wanted to save money and reduce latency and to avoid traffic tromboning over international circuits and back again. They did it because they didn't want to be screwed by the guy upstream from them. They wanted control over their own destiny. When you peer, you can have much more control over where your internet traffic comes to and goes from. Keith, talk about the opportunities in the future. As we evolve now and have years and years of experience running large scale networks. As the world changes, you start to see more of a global landscape certainly in the country by country basis on the US. Obviously it's here but in North America but outside of the North American in the EU for instance, obviously different Germany versus this. Data isn't our issue. So given the landscape of the global agenda, what are the opportunities that you see for entrepreneurs and for tech geeks because we were talking just before you came on about some of the hard challenges to solve these problems are out there. So there's certainly problems to be worked on and that will yield to opportunities. What do you see? Well, we have a huge concern in the internet industry while internet community at the moment about pervasive surveillance. And nation states and other actors have been able to get their hands in your traffic. Nation states saying, well we don't want our traffic to go via this other country because it might get snooped by that country and we don't trust their interests. By increasing the amount of peering and the way that it is understood, then you're never going to eliminate the bad stuff but you're reducing the risk surface. The other thing that's going on this week is the DEF CON conference in Vegas where all the hackers and all the good guys and bad guys get together. Internet security is a huge threat. I don't think that it's an insurmountable threat but we need to do things that tackle that and one of them is to have more control over our infrastructure and our traffic. Another is distribution of clue. Understanding best practice in internet security is really important. We need to do a lot more in terms of internet security and protecting the infrastructure. My organization, DNSO ARC, is very much about protecting the domain name infrastructure in the internet. I think the other thing is that when we figure out models that have made the internet successful in our western developed countries, it's also about helping advocate, not impose but advocate these models to other countries. A lot of governments in developing countries do not want to encourage competition in the telecoms market because they see the incumbent PDT operator as being a cash car. But actually there have been circumstances now whereby developing neutral internet exchanges at which local and country peering can happen that actually the cost burden of transit services to these countries, the cost of connecting to the rest of the internet has gone down significantly. And there are huge opportunities for spreading the knowledge and the clue that we have learned about how to do this stuff to make sure that the rest of the world, the other half of the planet's population that needs internet services can do so in a way that is effective and is not subject to vested interests of using it. And then you've got the internet of things factor, right? Which is coming fast and furious, which I don't know what your numbers are, but it's going to increase the devices, increase the potential, I like that phrase risk surface significantly. How much of that is a driver and how much of are you able, kind of as a standards committee organizations to get ahead of that curve? Well, to some extent it's a bit frustrating because the things that are desperately needed for the internet, things to work properly are things like IPv6 addressing, things like huge security practices, low latency networks, networks that are transparent across, not just the socket that's got your internet provider, but across your entire building, you know, whether that's your home or your business. So there are a lot of things that those of us who've been in the industry for a while have been advocating DNSSEC, IPv6, multicast. Many, many of these things are going to be absolutely critical in the security. All of these things are going to be really critical in the internet of things. I hope that we see the internet of things as an opportunity and motivation to go after all these solutions that have been looking for a problem for a while now. Keith, you had a discussion and you talk on the power of interconnection cooperation, I had to kind of read that from mouthful. Interconnection cooperation. Now just describe what that is. What does that mean? You can connect to one other party on the internet and you can both say, well, fine, we're gonna peer or we're gonna transit or I'm gonna sell you my local loop service or something like that. And that's fine if there's just two of you. And then if you start doing that relationship with three or four, then it starts to get very inefficient because you're duplicating the effort and the resources across each one of these relationships. When you set up something like an internet exchange or a clearing house for data about the internet's domain name infrastructure, then you're gaining from the power of many to many. You're building a community. Everybody is bringing their own contribution to the table. You get a critical mass there. Quite often a side effect to that critical mass is you finish up at the marketplace where people can buy and sell services from each other because it's no longer just a one-to-one relationship, it's a one-to-many relationship. What's your biggest thing that you want to share with folks out there that you've learned, that you see as a, or a question you get from folks that are either friends, customers, partners in terms of the peering, because peering is obviously a social thing as well as a technical thing. What's the number one thing that you hear from folks around the challenges, opportunities, and advice that you give? What's the number one item? That's an interesting question. Again, I think it relates to security and protecting infrastructure quite a lot, which is a little bit of a negative message, but as you see, it's a tractable problem. Yeah, you need to solve, that's a real issue. Yeah. How about governance? Does that hamstring the operation at all? I mean, obviously there's been very political, certainly on the governance side, and with the global footprint of DNS now, there's no one single country that runs it. Yeah, I think for governance, we've developed what I think is a 21st century model of running organizations, and I think a lot of the, shall we say, powers that be have not figured that out yet. And when people say, oh, we should have the ITU running the internet instead of ICANN, they're trying to apply a 20th century governance solution to a 21st, it should be the other way around. We should be modifying, we should be modernizing the ITU so that it is more like internet governance structures. I mean, I just, so much going on with DNS, I got to ask you the question, the CCTLD was a very big part of the infrastructure because the, you know, GTLDs were limited. So after now I just got an email that said, I could buy the domain John Furrier rocks, dot rocks, dot guru, I mean, I mean, what is coming, dot house? I mean, I mean, there's so many GTLDs. I was just reading last week that we are going to have the dot WTF and the dot fill, top level domain. So, yeah. Yeah, I mean, so does that mean the CCTLDs? I think it means that our governance processes are not perfect and we still need to work on them. But what does that do to the CCTLDs? Is it changing all the dynamics? It used to be that was a very specific country specific. Does it change it? Does it flatten it? What's your take on this? Well, I mean, something that's happened that's been quite controversial recently is, is dot UK, the nominate organization that I was involved in founding. They used to have like many versions of dot code dot UK for a British version of dot com and dot AC dot UK for a British version of edu and they've just flattened that all out. So you can get anything dot UK now. And I'm pretty clear that, you know, they have been motivated to do this because they see the top level flattening out. So, okay, we need to flatten out our national infrastructure as well. And yeah, I think that there's scope for opening all this stuff out, but I think that there are also dangers that are going too far. And I think, you know, the other thing to understand about the domain namespace from my perspective is that people don't type domain names into their browsers or email addresses anymore. They're using apps and search engines. That doesn't mean that the domain name services are relevant. It's just going to become another, and most people don't see IP addresses. In other five years, domain names will be an invisible layer in the protocol stack that people don't see, but is nonetheless completely mission critical to the internet. Which is why we built CrowdChat, which is going to be doing it for hashtags. So the new URL. Keith, thanks for coming on theCUBE. I really appreciate it. I want to give you the final word for this segment. Share with the folks out there what the vibe is for peer to what's it like here? Why is this inaugural event important, and what's your takeaway? I think we're seeing a great mixing of some of the up and coming talent in Silicon Valley that has made a real difference over the past decade. That has really made the internet, social media, the cloud part of everyone's everyday life. And we're seeing that intersecting with the internet secret source of peering that keeps the whole thing hanging together. So it's not just the usual faces here. It's new people and the new people interacting with the longer time people. And I think that's, you know what's- And setting the agenda for conversations to be discussed, right? Yeah, absolutely. Technical end business. Yeah, I agree. I well said. I mean, the people who have made it happen really giving back with their own time in the foundation to train and get educated, but also keep their eye on the prize, which is freedom, much more access, right? This is theCUBE. Thanks for coming on. I really appreciate it. It's a great perspective. We're going to the DNS. We're going to the international global landscape. This is theCUBE. Of course, we are flat. We're global. We just send the signal out and see where it goes. And I'm John Furrier with Jeff Frick here live in Silicon Valley in Palo Alto, for Pew.do. We'll be right back after the short break.