Max: 00:03 Welcome to the tech deep dive podcast where we let our inner nerd come out and have fun getting into the weeds on all things tech. At Clark Sys, we believe tech should make your life better, searching Google is a waste of time, and the right vendor is often one you haven’t heard of before. Hi. I’m Max Clark, and I’m talking with Bill Wanaka, who is VP of product management, global Internet and content delivery services for CenturyLink. Bill, thank you for joining.William: 00:27
Thanks, Max.
Max: 00:28 So, Bill, you have a a fascinating job to me and a fascinating history that goes all the way back to really the start of the Internet with Global Center. I’m gonna call you out there and how long you’ve been doing this. And have, you know, came into CenturyLink by way of, Level 3’s acquisition and have have really been at the core of the Internet for a a a long time now. I mean, just to kinda start here, I mean, when you look back and we we talk about, you know, a 20 plus year horizon of of time, what’s that been like? What’s, you know, I mean, how much has surprised you in that span, and and how much of it was expected as we saw this the growth of the Internet?
William: 01:03 You know, I I hate to say that, you know, I invented the Internet or something silly like that. But, but, you know, back in the Global Center days, we were working under a leadership team that had run May East and May West. So I got exposed to interconnection challenges pretty early on. And at that time, you know, marketing Internet services was really about overcoming the challenges of congestion on shared routers in May East, May West, and certainly other other important inter exchange locations, came forward. I I think if anything surprised me, honestly, it’s actually how long it took for the Internet to develop a deep interconnection, topology.
William: 01:40 And that the the, the the move to do that took as many decades as it did. Today, we’re incredibly deeply interconnected and incredibly resilient in North America. And then, of course, I think the rest of the regions in the world, don’t have the same benefit of homogeneity of culture and content. And so interconnection becomes increasingly more difficult as you leave the US. Certainly, it it works and it has a way of working, and there are certainly nuances by each region depending on, you know, again, those cultures and then also the separation of of, of of countries from each other.
William: 02:15 Right? In the US, it’s very easy to build deep interconnections if you have, you know, the right amount of traffic and the right resources. But as you get into the rest of the world, EMEA, APAC, and LATAM, it’s it’s really a very different story. So we’re really happy about the way the the US has evolved. I will say there were, you know, pain along the way.
William: 02:36 Certainly plenty of things in the public record to suggest that not everyone was getting along in every decade, but but now, you know, we are really have a resilient robust Internet infrastructure, and it’s showing its ability to withstand the strain of the increased, traffic to to the, COVID stay at home orders.
Max: 02:54 So the general population or public is not really aware of interconnection until you have a a public dispute. You know? And and that’s kinda the equivalent of DIRECTV canceling, you know, somebody’s streaming package. Right? But at the same time, I mean, it feels like capacity is always limited.
Max: 03:11 I mean, there’s there’s this this growth and this hunger related to interconnection and capacity just just doesn’t seem to really stop. I mean, is that really coming up in the US still? Or is that really a I mean, we see it with with, you know, APAC region networks. Europe seems to be a little bit easier in most cases. Lat LatAm is difficult as well.
Max: 03:31 Like, how how do you make plans around this, and how do you deal with that capacity requirement as this grows?
William: 03:38 You know, back back to what’s special about the US is that there’s so much geography, and the density of that geography is certainly concentrated in a couple of areas. But then as you start to fan out, there are a lot of dense populations through the middle of the country and on the coast that are, you know, are not the the original two cities, which were more or less San Jose and and Virginia, which is where the two interconnection points were. And so because the US has such a distributed population and because the peers of ours here that, you know, the the the training partners that we work with in the US, the largest mobile cable and and and wireline broadband suppliers, have a footprint that is broadly geographic. Interconnection happens in as few as 9 locations. And for some of our larger training partners, it happens in as many as 25 locations.
William: 04:26 But the fiber richness that, you know, we certainly have enjoyed, with the number of companies that have become a part of CenturyLink over the years, it’s it’s been, I think, an asset rich territory where interconnection is now considered a a priority for, user experience. And and we see some of the most aggressive training partners here really leaning in in the face of COVID to proactively upgrade in the case of one large training partner more than doubling our capacity between us in this period of time, that we’ve been in lockdown. And so the situation is very different in EMEA in particular. There’s been a lot of concern about congestion being caused, by various events that are happening. Certainly game popularity is increasing.
William: 05:13 I think all of our kids are probably playing Fortnite or Call of Duty at this point. And, that has created a lot of strain in in Europe in particular where, you know, we’ve seen inquiries, both formal and informal from regulatory, groups there and and and have had to, you know, take steps and measures to not just accelerate the capacity deployments in in those specific countries, but also, you know, we’ve been actually very busy writing software to control peak traffic and to help control peak traffic during business hours, which obviously, you know, congestion would have an impact on on on on commerce and the economy there, as well as, peak hours and peak viewing hours on the Internet, which are 5:5 PM to 11 PM.
Max: 05:54 I mean, do you think that the the issues there are related to the interconnection capacity being strained or the regional access networks having, you know, head you know, the equivalent of, like, a head unit, you know, at capacity and strain or the actual, you know, last mile into the subscriber location? I mean, what what is actually driving that concern or, you know, and and that capacity crunch?
William: 06:14 You made a really, really thoughtful statement earlier, which is there’s a finite amount of capacity. I’ll paraphrase it a bit, but there’s a finite amount of capacity in in any country, in any location. Interconnections in Europe tend to be concentrated within specific countries. And so as you as you start to look at what a big Internet event, you know, that sort of impact that a big Internet event can have in a single country. I mean, take Germany, France, Spain, or or or Great Britain.
William: 06:42 You could probably pretty quickly tally up just exactly how much capacity exists in the interconnection and down to the last mile. And over the last couple of years, we’ve had to really think hard about that because for some of the large events that we’re now starting to see in the Internet, whether they’re, you know, game downloads or whether it’s streaming video, the infrastructure has, you know, reached those points where you you actually start to see it, congest. It’s a little hard to pinpoint. It’s a little bit more about your weakest link is somewhere in that mix on any given day. But, you know, on the whole, there’s really a a finite amount of headroom that is there to be absorbed.
William: 07:23 And there’s lots of good ways to continue to grow and expand your scale in Europe. What you don’t have the benefit of is the same type of ability for a Great Britain to handle a Germany in the event of, of of of hitting, you know, large constraints in those countries. Well, while the carriers there do interconnect outside of their home countries, the capacity that they carry is, is very much oriented around actually us getting content into their edge and localizing it into their networks. And therefore, the backbone has become, you know, thinner than the edge. At this point, you know, our our edge growth is far exceeding our our our backbone capacity.
William: 08:02 So a catastrophic failure or any other pinch point that that could occur could make for a bad day in one of these countries. And and, you know, certainly, it doesn’t even take a failure at this point. The popularity of games and streaming video are, I think, pressing up against what’s possible in those countries. That’s why software has had to be introduced into the solution and into the mix to control session rate limits. So if I have a great broadband connection, you have a great broadband connection, we live in the same country, in the same city, we’re not gonna get the same service, obviously.
William: 08:35 So so, finding ways to democratize and flatten that not only, across users on the endpoint who are demanding the concept, but also across the suppliers of the content has become really important in this event.
Max: 08:47 So, I mean, I hate to use the term because it’s been on a huge marketing hype cycle for a while. But, I mean, isn’t this really the play for Edge and why Edge is becoming such a big thing in conversation about getting content and services closer to the subscriber and beyond the interconnection locations? Like, you know, how, you know, how does this affect Ed’s strategy, you know, for CenturyLink and and when you’re looking at at product planning and and and service delivery?
William: 09:12 That’s a great question. You know, in in the content delivery world, we’re primarily using, HTTP protocol to deliver content to consumers on a mobile device, on a connected television set, on a on a laptop. I think increasingly, though, we’re seeing really strong competition from the over the top videos folks, the the esports folks, and the gaming companies. And so what they’ve needed from us, and I think what they need from us going forward and the big trend that we see is that they will demand to have more flexibility in that edge. So I can tell you right now, you know, we’re already working on releasing a product in the Q4 that will give a compliment to our content delivery network in the form of a bare metal as a service.
William: 09:58 And additionally, beyond that, we’ll be launching various layers of virtualized services on top of it. And one of one of our, very, very aggressive customers is, you know, acquired proprietary streaming protocol, platform that enhances user experience and also enhances the ability to monetize the content. And that is not something that can run on a conventional CDN content delivery network edge. It’s, you know, it’s it’s it’s complimentary to those services that still need to serve all the devices we normally serve, but they are absolutely striving to deliver a differentiated user experience. And to do that, you know, they’re using proprietary software.
William: 10:38 So we’re looking at not just enterprise applications there because there’s a number of enterprise applications where the edge can be used to reduce the congestion that’s sent upstream, as you say, collecting video security or surveillance footage from from a location. You know, edge compute can be used to reduce that data set before it has to go back to a core cloud service. But in the case of the, you know, the entertainment industry, they are absolutely focused on finding ways to compete and, win, you know, user user attention, by delivering better experiences. So we’re seeing a lot of innovation there from our largest gaming customers and our largest OTT customers.
Max: 11:16 So, I mean, a criticism of CenturyLink would be that you guys were are are slow moving because you’re so large. Right? And a lot of other CDNs have come up and become very specialized. And what you’re talking about really is, you know, customers that need a very specialized and unique experience or service tailored to their exact business model or service delivery model that can’t be accomplished by a generic CDN that can service everyone. And how do you actually partition out your platform?
Max: 11:43 You know, I mean, there’s other global players, but you seem like you have a lot of unique, you know, assets and abilities to actually take your scale and your leverage in in deliver a unique product and experience for for people looking for these things.
William: 11:58 Well, thanks for that. I I think that that’s a back hat and a compliment, and I appreciate it. I I guess what I’d say is this is an opportunity for for me to declare that we have, looked at our network assets and we’ve looked at what’s happening in, you know, both more traditional enterprise space with web applications and, and industrial grade computing needs. Meaning, I need to have a location where I can run workloads that is, from a latency standpoint, within 5 milliseconds of, the an a user location, whether that’s a manufacturing location or it’s a retail location or any type of enterprise location where latency actually dictates the ability to run the workload. And, you know, the cost of the cost of running a workload in an enterprise location is high relative to break fix, relative to the ability to service and manage, you know, the equipment on on a day to day basis.
William: 12:50 And so as we looked at our network and we looked at our on net locations in the US, we were we were in a position to provide edge computing services that really no one else was within, you know, 5 milliseconds of user locations on our network and covering most of the population of the US most of the business centers. We’ve pivoted, I think pretty quickly toward enabling industrial use applications across our private networks, our IP VPN, our SD WAN, products and services. And, on the internet facing side, we also enjoy this series similar low latency and and, you know, high degree of ability to put content very close to where the users are. We have over a 120 pops in our CDN today, and that that’s actually growing, not declining. We’re we’re really trying to ensure that localization of those services can enable these types of new applications.
William: 13:48 And I’ll give you some examples. You know, tactile application like, augmented reality will absolutely require sub 10 millisecond latency, any sort of industrial or manufacturing controls or or sub 5 millisecond latency, types of applications. So the pivot here has to do with the fact that, you know, we looked very closely at our assets and our products and decided that we were going to need to become a technology innovator if we were gonna take advantage of the physical network. So things in the physical world move somewhat slow relative to the software world. So you’ve seen us do an acquisition just recently in September, required a a company out of Paris, called called Stream Root, and they make a technology that allows for the actual sharing of popular content.
William: 14:33 So if you and I are watching the same show at the same time, and we’re sitting in here in the same market on our broadband connections. The software is clever enough to determine whether you and I are watching the same content. And if we are, can the network be more efficient by downloading chunks of that content to to each of us and reducing the amount of downstream bandwidth that’s required. This type of technology works incredibly well behind a firewall in an enterprise location. And in fact, you know, we have several customers in the, UCC space that are deploying this technology inside of their applications to improve high quality bandwidth, in locations that are hard to reach is the the term I’ve been using.
William: 15:14 Sounds like a place in your back that you can’t quite scratch, but, but that is the case. You know, downstream constraints in in enterprise, you know, when everyone wants to watch a high quality video of the CEO, talking about, you know, the state of the company, those things have traditionally delivered very poor video experiences. So, you know, I think, again, we’re we’re really looking for the assets to give us an advantage. I don’t think that there’s this, you know, we we have an amazingly unique set of fiber assets. We obviously have several layers of network services above that.
William: 15:45 And and and now we’re really looking for software that can help us accelerate, you know, the ability for customers to to consume that that asset.
Max: 15:53 And I think you you defined edge here very interestingly, which is, you know, within 5 milliseconds. A lot of the original edge hype related to installing equipment at the base of a cell phone tower and, you know, building out all these little mini data centers and shipping containers or something. And that that has not proven to be an effective deployment of edge. But when you look at 5 milliseconds from a network asset, that’s a pretty big distance. You know?
Max: 16:17 And, I mean, in terms of mileage, I mean, that’s that’s a a relatively large metro area that you can cover in 5 milliseconds.
William: 16:23 That’s right. Yeah. So it’s it’s a large metro area. There’s a number of players in the space that are focusing on the, radio access networks and supplying, bare metal as a service and other virtualized, services to enable, you know, virtualized network functions. So we think that that’s a really, really good trend.
William: 16:42 We do do fiber in the metro for for mobile traffic backhaul. But we haven’t jumped out and entered the space for for mobile, you know, carrier, grade edge compute services. Our footprint is, I think, well suited to enterprise use cases and applications. We’re certainly talking to all of the organizations that are trying to do the, you know, the mobile cell tower enablement. And I think it’s gonna be a very complimentary relationship where we’ll be the next hop back from those providers.
William: 17:13 You know, you can think of the the popular providers in that space and and assume that they’re going to be using our fiber and our assets, and they’re going to want to have interconnection and probably more importantly, breakout of layer 3 IP traffic. It’s sort of silly to think that we’re going to localize some of these applications, possibly as close as within the metro and then fall back to, you know, even 9 locations in the US to actually exchange the, the, the layer 3 internet traffic. As much as those interconnections in the US today are, you know, in somewhere between 9 25 markets, I do see a world where we’re actually interconnecting in all metros and we’ve created a much more scalable Internet. So edge for us at that point becomes we are sitting as the next hop back from the aggregator who’s delivering the mobile mobile tower edge compute. They do not expect to have enterprise or consumer applications running there.
William: 18:04 The space and the power is far too limited. Max, I guess the other thing that I think is going on is anytime you have a new space like this, there’s a lot of hype. What was important for us was to take the time and really look at the assets and determine if we had a sustainable competitive advantage here. And I think the other way that we’re capitalized, and, you know, our our profitability and our ability to invest in the future, those are the things that are gonna differentiate us from the startups that have tried to enter the space or or those companies whose physical footprint and assets don’t match, the type of deep localization we’re talking about. Well,
Max: 18:36 I mean, you’ve got decades of head start and billions invested in this. And you are the largest network on the planet by every survey that comes out. I mean, it’s you know, so you should be able to leverage that weight and and get something out of it. So COVID, has driven a massive shift in workforce. You know, we’ve seen, you know, this what was a slow uptake in remote or distributed working environments go all remote overnight, and also at the same time, a fantastic increase in residential demand in, you know, other services, you know, telemedicine, you know, schooling, video gaming, etcetera.
Max: 19:11 So I understand if you can’t answer this specifically for CenturyLink. But for a, you know, industry, I mean, what kind of numbers are we really talking about here in terms of growth and capacity demand in the last month as it relates to, you know, what has now become this event?
William: 19:24 We’ve seen a 50% increase in billable traffic daily since this around the time of stay at home orders. It it’s kind of a perfect storm as well. You have several new products that were being launched, certainly innovations in the online gaming space. I think the demand from UCC and and and video conferencing is is obvious to all of us, and that has hit all verticals. We’ve also seen a number of new OTT products hit the market in this period of time.
William: 19:51 We happen to be in a pretty good position ourselves because we were anticipating the launch of the new OTT products and we were anticipating the launch of the gaming products. So we had, we have been in a pretty heavy investment cycle leading up to this. And again, I think that the cooperation and collaboration with the carriers and our trading partners globally has just been outstanding. We’ve really seen a lot of willingness for everyone to come together and work to ensure that we’re reducing the impact of, of of of the different types of traffic that are are potentially, you know, impacting. I think at the start, you know, did impact in in many cases, Internet performance.
William: 20:31 So it’s every area of demand that, you know, you you could think of relative to Internet access and and internet content delivery has seen this type of, of really, you know, radical overnight growth. But again, where there have been hotspots, there’s just been a really strong willingness of the ISPs. And, you know, in case of game downloads, the game publishers and game distributors and CDNs to work together, more transparency than we would probably have if we didn’t have a global pandemic occurring. And certainly there’s been encouragement to lean in on on, aligning practices so that we don’t have 1 or 2 bad actors that are out there spoiling it for everyone else. And I’m talking, you know, daily phone calls with, you know, multiple customers and multiple ISPs, that you know, could go on for a week or 2 at the start.
William: 21:20 But in general, whether we were determining where and how to deploy capacity, improve the situation, how to shift traffic around to improve the situation, or in many cases for us, writing software to help some of these customers that didn’t have controls in their platforms to control that peak, that that that actual spike when everybody wants to get the same video or everybody wants us to get the same game that just, you know, just blows up all the other traffic. That that has been a phenomenal story. And you’ve seen them in the press. You know, certainly, a number of the major OTT providers came out and and mentioned that they were taking those steps. Obviously, we’re the company behind a lot of those those customers, and then you’ve seen it in the gaming space as well.
Max: 21:58 So it it seems like, I think it was somewhere towards mid mid to end 2018, early 2019 that the cost of running 100 gig interconnects, you know, the economics equalized or it got improved over 10 gig legs. And I mean, you know, this is a factor of how much does it cost for the router and the blade and the optic, plus how much power does the optic suck in order to, you know, deliver these circuits. And since then, there’s been this explosion of 100 gig interconnection, and it seems to be accelerating rapidly. I mean, this goes back to an earlier comment you said about, you know, capacity and and constraints. And as we look forward now, you know, there’s always, you know, some some manufacturer, you know, political battles related to what the next spec is gonna be.
Max: 22:39 Is it 200 gig Ethernet? Is it 400 gig Ethernet? Is it faster than 400 gig Ethernet? How do you plan for that? And and and where are you in the planning cycle of, okay, now we’ve got, you know, x by 100 gig legs deployed at, you know, as our interconnection, but we already need more.
Max: 22:55 And and how long does it take to get ahead of that curve, and and where are you guys in that cycle?
William: 22:59 I mean, I think 100 gig ports are are in their sweet spot at this point in terms of economics. And we all will always, insist on 100 gig ports unless there just happens to be no other option for our interconnection partners. From a customer standpoint, I think, you know, 10 gig is going to have a long shelf life. And certainly, you know, our backbone is is made up of a lot of 10 gig customer ports and now actually, you know, thousands of 100 gig ports from the customer side. But on the interconnection side, you know, we really drive towards towards all 100 gig interconnection.
William: 23:31 There’s another project that we’re working on internally to bring the largest carriers and cloud providers together in the US to interconnect at terawatt terabit speeds. And some of the thinking there is that by building a switching fabric that everybody can access in these metros, that we can actually accelerate the rate of turning up, you know, the interconnection services. So the largest cloud providers and the largest, interconnection partners and training partners, have come together. We’ve we’ve kind of led that charge, and we’re building out optical platforms now that, will be standing in place with, you know, minimum upgrade capacities of terabit at a time. So we are planning for an even greater acceleration of traffic exchange between content providers, cloud providers, and and eyeball networks, and playing playing a pretty central role in in getting that project off the ground.
William: 24:25 It has not been announced yet, but it’s, you know, it’s absolutely, I think, leaning into the future. But to answer your question directly, I mean, a 100 gig line speeds are are what make up those terabit connections. So, you know, the the focus is still on where the economics are the best. And then certainly, you know, we tend to believe in the, multiples of 10. That’s just, I don’t know if that’s been circumstance or not, but it feels like the interim steps that happen, you know, the 40 gigs, the 400 gig steps, those are those haven’t ever really been, something that we could justify economically in making the change.
William: 24:58 So we tend to go on the cycles of 10x multiples here.
Max: 25:01 Well, I mean, you you said a 120 pops for, I think, just a CDN. And when you look at that times the number of routers and devices you have across those pops, like, all of a sudden, you’ve got quite a large fleet of equipment that you have to maintain and then upgrade in those cycles.
William: 25:15 Yeah. That’s that’s absolutely true. The goal, I I think for us is to continue to expand the network into the places where customers wanna go. So if I you know, I mentioned before that we had done a lot of pre building before we hit COVID, and it had to do not with the fact we anticipated COVID. Obviously, no one can anticipate a black swan event.
William: 25:34 Otherwise, it’s not a black swan event. But the fact of the matter was is that because of, our investment position and our ability to spend capital in advance of customer demand. We went to the places where customers said they needed us to go. And I think that’s something you can always come back to Central. I can say, if I’ve got, you know, if I’ve got a critical initiative to digitally transform my company, Who are the players in the market that I can go to?
William: 26:03 Show them what I’m gonna do and and convince them that, you know, it’s it’s in our mutual interest to build. Those are the projects that we love. And I think that’s why we get the crack at some of the biggest projects in the world is because, you know, we we, we have that appetite to build where our customers need us. And whether that’s a single enterprise location and us bringing fiber into that location to ensure it’s continuously upgradable from 10 gig to a 100 gig to a terabit. Those are bets that we take with you all day long, and, those are the bets that our company successes is predicated on.
Max: 26:36 Hi. I’m Max Clark, and you’re listening to the tech deep dive podcast. At Clark says we believe tech should make your life better. Searching Google is a waste of time, and the right vendor is often one you haven’t heard of before. With thousands of negotiated contracts, ClarkSys has helped hundreds of businesses source and implement the right tech at the right price.
Max: 26:51 You’re looking for a new vendor and wanna have peace of mind knowing you’ve made the right decision? Visit us at clarksys.com to schedule an intro call. I mean, CenturyLink has been building like crazy in like APAC, and there’s a couple interesting things going on in the APAC region. But, I mean, one of them is it it just doesn’t seem like there can ever be enough capacity. I mean, you’ve got almost unlimited demand.
Max: 27:12 And at the same time, as soon as you upgrade capacity, it’s it’s almost entirely saturated. I I I how how much of that is just this frustration cycle? And and and how do you how do you make plans for that and keep up with it and and balance, you know, the financial cost of those investments and upgrades as well as with your trading partners? I mean, you’ve got 2 people here that have to make decisions and come to an agreement to collectively invest a lot of money to, you know, to to increase capacity.
William: 27:37 Supply chain has certainly become an issue that I don’t know that any of us, again, were prepared for in the sense that supply chain intervals have gone in some cases for some types of equipment for some of the competitors. And I think for for all of us, you know, have have extended to more than 2 x what they were before. Building an Asia Pac is particularly hard because of the lack of cultural homogeneity and because of the water that separates some of the most important countries over there for US customers. The other thing about building, you know, this is funny because it’s such a chicken and an egg. It’s it’s a question of if you have the traffic that needs to be delivered, there’ll be a trading partner who is there to receive that traffic.
William: 28:20 And those negotiations tend to go more quickly and are usually financially neutral for both parties. When when you don’t have the traffic, everything is, you know, starting from scratch and trying to build from the ground up. And and so the the the economics of the business really demand that you have to have the traffic in order to actually build the scale that’s required to handle, you know, industrial grade or consumer applications. And so we did do a big lift in Asia Pac. We announced that at the end of last year and, it was driven on the back of our conviction that the customers we were working with were going to have a product that have products that were extremely popular in those countries.
William: 29:00 So we looked at their roadmap and their build plans, and we actually frankly started building prior to some of those even emerging. But again, looking at the way that these US customers wanted to enter that region, we were able to determine what countries would be most likely to be prioritized for them, start that building process. And and as, it’s just as you said, pretty much as soon as we open up the ports and started opening up the traffic, both through our Internet backbone and to our content delivery network, the traffic just flowed. And then again, once the traffic’s flowing, it’s a whole different game. You can actually go much, much faster if you have the balance of traffic.
William: 29:37 That’s that’s when people have to take you seriously. So it’s been a really rich and rewarding experience, but the supply chain, you know, fears are are real. We’re not particularly dependent on any one country or any one vendor. We’re extremely diverse in our use of multiple vendors for for lots of different reasons. And so that’s allowed us to be, I think, pretty nimble in the face of of what for others might might be supply chain issues.
William: 30:04 That and also having the deep end strength and, you know, the ability to invest capital. And then what we’ve seen is as traffic has started to go into countries that, you know, weren’t in our original build plan, we’ve had a lot of our carrier partners in these countries come forward and ask us to expand more quickly into their countries to reduce, you know, the burden of subsea, backhaul and capacity challenges that they face. And so we’re taking those investment opportunities incredibly seriously and putting capital in flight to, I think, improve user experience in those countries and ensure that those networks are operating more healthily as well. It’s in our I think it’s in our interest to help any other carrier who’s come to us and said, I’m seeing so much traffic from you in this country. We’re interconnected there, but I’ve got to back haul it across the ocean and and deliver it into, you know, my eyeballs over here.
William: 30:53 So that’s created a lot of really interesting and compelling opportunities for us to accelerate that growth now that the traffic’s there. And, again, there’s there’s lift. There’s no question there’s daily consumption lift from from COVID as well. It’s while we’ve we’ve been very fortunate to have the right customers and pick to bet on the right customers and grow with the right customers, we are also seeing a general lift in traffic that is absolutely based on the increase in consumption. That’s created new opportunities for us to to lean in even heavier in Asia Pac.
Max: 31:19 How difficult I mean, so the Chinese market is a massive market for companies. And in a couple different ways, you know, companies that need to get access to the Chinese market or get access to Chinese eyeballs and vice versa, you know, Chinese eye you know? But at the same time, you’re you’re dealing with a with a I don’t know what to call it. A regulatory I mean, it’s more like a geopolitical, you know, becomes a driving factor in these things. And based on rhetoric in the press or on social media now, I mean, that has pretty wide ranging, you know, impact in your business and your planning.
Max: 31:50 How do you I mean, is it just keep your head down and kind of ignore it and just keep moving along and just know that these disruptions are gonna happen or you, you know, can get derailed? I mean, how do you how do you maintain and manage that when there’s, you know, this this additional layer of complexity that can come out of, you know, nowhere all of a sudden?
William: 32:07 China is a very unique place in the world relative to interconnection. I’ll start there, and maybe we’ll go to supply chain after that. But it’s a very unique location in the world due to the fact that the government is using, essentially a large firewall and that all of the carriers implemented to enter and exit the country. And then operating there is incredibly complicated. When I say that, I mean, it’s very different from operating in the rest of the countries in Asia Pac or anywhere else in the world.
William: 32:33 So we tended to use interconnection outside of the country to ensure that we’re delivering, sufficient access from our network into China and then from China back to our network. And then we use partners in the country to do last mile Internet connectivity and broadband, and that has worked extremely well for us. I think the complexity of operating there, given all the other complexities and challenges that we can go tackle, you know, outweighs the benefit at this time for us to go in and try to establish a business entity locally. That’s definitely on the list of places we’re going, but it’s, it’s not the priority. I I could I could certainly point to other markets in in Asia, Pat, that we wanna, continue to grow scale and and more quickly.
William: 33:16 I think being tied to China from a supply chain standpoint has, some significant drawbacks at the moment and and has for for a long time. That’s that’s been well covered in the press, and I think everyone knows what we’d be talking about there. But it’s the last place I’d wanna be is dependent on China for my supply chain. And And again, that’s why the diversity of vendors that we’ve employed in our back, in our networks, allows us to have pretty good resiliency in an event like this where I think it’s pretty clear that if you if you were dependent on tech from China at this point, you would have seen some slowdowns and extension of intervals in supply chain. Yeah.
William: 33:53 It’s an incredibly important place to be. We work with partners there. We’ll continue to work with partners there. We’re about to bring a CDN partner on in China. Most of our competitors in the CDN space are generally reselling services that can be acquired at some level of effort, and we wanna ease that for our customer base.
William: 34:11 So China is on our road map from a CDN perspective as using a partner and country to deliver. And I’d also say that, you know, the streamer technology that we acquired has incredible applications in a location like China, where our ability to use that software technology, you know, it’s going to be a real advantage for for anyone who decides to take advantage of that. The tech is the technology itself and this this type of peer assisted delivery has, caught in a lot of attention and and actually gained quite a bit of traction inside of China as well. So it’s generally regarded as a good network traffic control tactic to deploy. So we’re excited about the opportunity for our technology there.
Max: 34:51 So, I mean, so so CenturyLink, you you go on the all the way to the bottom. Right? You have fiber in the ground or on poles or in the water, and then a network’s built on top of that. And then services are are built on top of that network. And, you know, we’ve talked you know, hit, you know, briefly on Internet or MPLS or SD WAN or content delivery and these sorts of things.
Max: 35:07 But there’s a lot of additional stuff that you guys do that people don’t really talk about or know about. I mean, you see a massive swath of Internet traffic on a daily basis. You have very popular recursive name servers that are I I couldn’t even I don’t even know the scale of how many queries they’re doing per day. And then on top of that, you layer, you know, a threat intelligence product and, denial service mitigations, you know, product, and you’ve got ancillary security services. You know, these all kind of work in harmony, but have have competing interests.
Max: 35:36 You know, the need and interest of of content delivery and rapid interconnection and and capacity is different from, okay, we’ve just now enabled additional ports and put more people on that can DDoS somebody. I mean, how do you balance that? What does that look like? And how, you know, how competing are these objectives for the business with each other as you scale and continue to grow this?
William: 35:57 One of the things I’ve always liked about working, at CenturyLink, and, you know, I’ve got a 22 year run through acquired company at level 3, is that, our CEO is very focused on making sure that decisions can be pushed to the right level of employee. And I’ll just say this, all of those products that you just covered, my peers and I all sit underneath one executive product leader. And, you know, we meet weekly, we meet daily, our projects, intersect and intertwine. For example, I’ve just launched a product called CDN Edge Compute, which allows us to deliver web application firewalls and bot management, bot detection. That’s actually stuff that sits in, you know, in my peer on the security side, Chris Smith, that sits in, you know, in in in his his his realm as well.
William: 36:41 So we launched this product together with the DDoS. It’s a it’s actually a bundle of services that allows us to write application layer protection as well as, volumetric DDoS mitigation. And I think one of the things that helps that happen is that we have an incredibly tight management team that’s worked together for almost 20 plus years and, has in many cases, all of us have worked on every product at some point in time. So there’s a strong, I think, awareness and and frankly, desire to help each other be successful. And I don’t know that you see that in every large company.
William: 37:14 I feel sometimes like Jeff’s Jeff’s story, our CEO, his his his desire to push that decision making down the levels where real change can be made is is is what’s enabling us to go faster now as as the new CenturyLink. And, again, I’d say, you know, with all the products we could talk about, there’s probably hundreds of products in that portfolio that is run by a team of about, you know, 8 product managers that report to 1 person. So our interests are incredibly aligned. And as these products start to come together, you think about SD WAN and its reliance on the security side of the business. You think about CDN and its reliance on the security side of the business.
William: 37:50 You think about our voice services and their suitability to our SD WAN products. These products are coming closer together, and I think our goal is to really, you know, drive towards a a single platform for consumption, especially as we’ve looked at virtualization as being absolutely critical to getting new products and services out quickly into the network. The whole edge compute story takes on such a, you know, an important central function for all of the product managers to rally around. So, you know, as we’re looking to transform customer experience, I think our goal is to make the products and services we sell easier to consume for customers and remove the friction that sometimes exists when you have multiple products that were not contemplated in the same, you know, the same organization. So our our go forward our go forward is very much about using a common core platform for CPE, a common core platform for edge compute, and then landing virtualized services on top of that that can be connected with software.
William: 38:47 And that’s where we’re spending our time.
Max: 38:49 So as you look at, you know, you say common edge platform for edge compute and the ability to actually launch containers and services for customers on the edge, I mean, that that puts you into, you know, the the cloud space. I mean, you know, public clouds are very large customers for you. You know, does that create a conflict at some point? I mean, is there you know, is that even perceived as, you know, competitive? I mean, given their size and your size in the market as it relates to compute and and capacity.
William: 39:13 We’re obviously a critical supplier to all of the major cloud providers, and we also play a role in managed services where we support, you know, cloud migrations. So we’ve actually also become a very critical partner to those major cloud suppliers. They’re certified from multiple platforms to handle migration as well as application development. You know, the relationship that we have with them is so multifaceted. You take any of the large hyperscalers and and and you look and you see that we’re involved in their consumer, delivery of their consumer services.
William: 39:46 We’re involved in the core infrastructure for their computing as a service platforms. We are involved in helping them deploy their edge capabilities into our footprint. So we’ve talked about our edge compute services going from bare metal as a service up to multiple layers of virtualized services. But we’re also pursuing the ability to allow customers to take and put those those core cloud companies, edge offerings into our data center. So a big part of our edge strategy is a lift on colo that would enable us to support the, you know, the proprietary edge cloud services of the largest hyperscalers.
William: 40:22 And and then having the certifications to do that transformation, you know, I think makes us an an even more valuable partner to them. Our fortunes are very tied to, the success of cloud as a business, and I think it’s become imperative that we get off the sideline. It’s one of the really nice things having come from level 3 in CenturyLink is that, you know, I’m part of a part of a cloud team. The cloud product manager is, sitting right next to me in those in those same meetings. And so, you know, we all look to him for sort of funny.
William: 40:50 You’ve got the the fiber and the colocation guy, and he’s at the first layer of our of our of our of our, of our stack here. And then, you know, you’ve got the next product management function, which is coming from our, edge cloud product team and core cloud product team. And the rest of us at the application layer are all dependent on that guy and on the, on myself on the internet guy. So it’s, Yeah, we, we, I think we’ve really reoriented the company to become not just a cloud networking company, but a a company that can use its assets to to enable customers really at any layer that they need to be.
Max: 41:27 About 10 years ago, there were some financial personalities on television related to stock market predicting that, you know, data centers were dead and the cloud is the way and get rid of the data centers. And and it was funny to me at the time because, you know, this misconception that, you know, the clouds were in data centers and you needed the data centers for the clouds to exist, but just really the perception. And it it feels like at the same time, you know, fiber and the network like, nobody’s cared for a long time. And now all of a sudden, it feels like people appreciate and care about the network experience and are seeing the impact of having strong networks and strong interconnection and strong you know, and and capacity. And it’s gone from this invisible thing to a very visible forefront conversation of, okay.
Max: 42:08 Great. We have our application. It’s sitting somewhere. How do we connect to it? What happens when tens of thousands of people have to work from home and access whatever platform it is?
Max: 42:16 And what does that look like? And how do we manage those things? And and I think IT departments are scrambling really fast trying to now rethink a lot of application connectivity and delivery. You know? So forget just just public Internet, but even also internal Internet.
Max: 42:30 And and I feel like, you know, you guys have some interesting stories that if you guys, you know, went out and and and shouted from the rooftops a little bit more, you know, you could really take advantage and benefit a lot from us.
William: 42:42 Yeah. I you know, we might be one of the best kept secrets. Certainly there’s been a lot of focus and attention on 5 gs as being where edge and edge cloud is happening. But 5 gs is incredibly dependent on power and space and Colo, incredibly dependent on fiber and interconnection. You know, you get off the radio network pretty quick.
William: 43:02 And certainly with this work at home situation, it feels like broadband has taken on a whole new importance. Just kind of what you said about data center versus cloud, Home broadband is where I imagine much of this traffic is concentrated right now. Yeah. Putting connectivity services is funny. We we launched the service, and I’m I’m sure you’re familiar with the Cloud Connect several years ago.
William: 43:25 It was generally, you know, a a data center to cloud operator product where customers were using that to migrate applications to the cloud, or they were using that for load balancing of their other applications between the cloud and physical data center. And these connections were really very physical in the past, but, you know, we’ve evolved that that stack to become completely virtualized. And I thought one of the most interesting use cases, while it’s still data center to data center, is that that service now enables us to turn up dedicated connectivity between 2 instances of AWS cloud services. And so rather than using the routing that AWS provides by default, which is, you know, let’s call it the default routing, We’ve put premium service into the market that allows a customer to virtually instantiate a connection across our network between 2 AWS availability zones. We’re gonna be extending that capability to support multi cloud.
William: 44:18 And I, I think what it comes down to is you’re, you’re right in the sense that customers aren’t gonna wanna have to deal with the physical much longer. Physical still matters. And where physical can be used to provide a diff you know, differentiated service or a service that stands out because it meets some specific set of requirements, that has to be software activated now. That those are table stakes. I mean, I do think that we’re moving away from the days when enterprises wanted to employ a data center manager and that there was a, you know, a buyer a buyer of Bandwidth who who, you know, dealt with the long lead times that were required to get physical circuits in place.
William: 44:51 The country’s really well fibrored now and the game is moving to software defined capabilities. And that is absolutely where the focus is in terms of our of our core of our core networking capabilities. SDN, software defined networking, is really where the investments are going. And the goal is to enable enterprises to be much more nimble in their ability to acquire a quality of service that’s appropriate for the task and acquire that instantaneously. And, that’s the challenge we’re that we’re rising to.
Max: 45:27 So when we look at global Internet, interconnection, relationships, management, etcetera, I mean, how much of this is science and how much of this is soft skills and relationships? And when you look at the soft skills and relationships, you know, sometime in the next 30 years, you’re probably gonna wanna stop doing this and go retire and hang out on a boat somewhere. Right? But how do you develop, you know, your staff working for you and and and bring up this next, you know, next generation of leaders that are gonna be driving the charge and managing, you know, these networks and and these and and planning going forward?
William: 45:59 Well, your question makes me feel a little bit like a dinosaur. But I will say this, I I think what’s, you know, incredibly important here, first of all, is that we make we make the transformation. Acquisitions often are about the product, but in many cases, your acquisitions are also about the acqui hire and being able to bring, you know, innovative and creative minds into the company and people that are going to fulfill really strong roles. You know, the organization that I’m in today is really thinking hard about how to include how to increase diversity within the workforce. And to me that that includes, you know, gender, race and age.
William: 46:38 So absolutely getting a diversity of backgrounds and skill sets to the table is critical. You know, from a customer standpoint, I think it’s as difficult as ever to navigate the number of choices out there. I really believe that you, for example, play an absolutely critical role in, helping customers to navigate what is an otherwise very, very confusing part of the market. And it is hard to tell where cloud blurs with network at this point. Chicken is why, you know, we’ve tried to make this transition that I described, earlier.
William: 47:11 But I think it is very challenging now for an enterprise to to to go to the next step unless they have somebody on their side that is very familiar with us and has the relationships with us and has the ability to reach out and, you know, get a conversation with our executive team. At this point, for most enterprises, they they don’t have dedicated resources that are doing that anymore and expect to be able to do it all virtually. But at the end of the day, it’s a big world with a lot of vendors and and a very confusing marketplace.
Max: 47:39 So we’ve been through, you know, periods of consolidation and expansion, startups, and consolidation and bankruptcy. You know, you look at the Internet and the and the providers in general. I mean, what’s your forecast for the next 5 years here? I mean, we’ve got a a massive amount of capacity being required with with with COVID, likely a pretty significant change in how people work as a result of COVID. I mean, what are you forecasting, and what are you guys getting ready for in your next cycle?
William: 48:01 Well, we’re we’re getting ready for, our customers to start to preference their digital distribution channels over their physical. That that that’s a pretty there’s a pretty clear line of sight to that. We’re we’re certainly cognizant of the fact that, over the top content will continue to grow at a more rapid pace. The online gaming space is absolutely heating up, as as we’ve said before. And I think that that will continue to morph into more new experiences that are delivered to consumers, which will require more flexibility on the edge.
William: 48:33 Security for work at home, employees of of enterprises is absolutely going to be critical. We do expect to see a major shift in the way that companies, view their real estate investments and that this will have a sustained transformation in the way that we work. So the dependence the dependence on tools to to enable those sort of trends, which are were there always, but accelerating at a rate that was, slower than I think, you know, what what we’re experiencing now. This is, it’s going to be a a very quick acceleration in the cycle of those those trends that were already underway.
Max: 49:13 You say dinosaur. I mean, I got on the Internet with a 1200 baud modem. Now I’ll pretend like I was 3 when that happened, but, you know, it’s it’s it’s it’s amazing to to to think back at, you know, where this started and where we are now and just the scale of it and just how quickly it keeps growing and how that pace, you know, continues to increase. Bill, any I’ll I’ll give you the closing thought here. Anything we haven’t touched on or you wanna talk about?
William: 49:37 I think that the human impact of this of this global pandemic is is just incredible. I know I’m stating the obvious there. We’re very fortunate essentially to be able to play a role in maintaining high performance on the Internet throughout this, event. You know, our services, are sitting behind the many, many, many of the UCC tools that have become critical. Our internet backbone or content delivery network have been integral in helping people stay sane in this process.
William: 50:09 And I think we just were very fortunate to have an opportunity to play a role in helping commerce to go forward through this event and helping, families to stay entertained and not go crazy on each other at home, but nothing to do. I just I just feel very fortunate that that I was here to be a part of this and that my company was here to be a part of it.
Max: 50:28 Awesome. Bill, it’s always a pleasure speaking with you.
William: 50:30 Thank you, Max.
Max: 50:32 Thanks for joining the Tech Deep Dive podcast. At Clarkstas, we believe tech should make your life better. Searching Google is a waste of time, and the right vendor is often one you haven’t heard of before. We can help you buy the right tech for your business. Visit us at clarksys.com to schedule an intro call.
