Digital Realty Trust, Inc. (DLR) Earnings Call Transcript & Summary

So with that said, I'd like to introduce Chris Sharp. And maybe, Chris, you could just do a quick couple of minute intro on yourself and we can go from there.

Absolutely. And thank you, Cole, and it's great to be a part of Edgecon. Unfortunately, this is my first attendance virtually, but always -- the calendars are a bit tough, but I've always wanted to go, and I've heard great things out of this group. And so I am excited to be a part of it because it's very -- it's not very often you get to talk about the Edge, where you don't have to spend 20 minutes about what the Edge is and that there is many different permutations to the Edge. So it's nice to be a part of this group. So appreciate that. I am the CTO of Digital Realty, where we have 275 data centers all around the globe. And most recently, at our marketplace live event, we talked about PlatformDIGITAL. And launching PlatformDIGITAL, which I think is a unique way to represent the value that customers can get out of our infrastructure. And this morning, we have the fortunate ability to announce some work that we had talked about at marketplace live, but really start to formulate more of a detailed plan around our piece of concepts in 3 markets, which we think we'll go through in more detail in our presentation. But I want to echo something that Cole said, and it's driven career decisions in my life that it's put me to where I stand today with Digital Realty, and that is about being open, right? And I think that openness allows for innovation in ways that we don't necessarily see today. But I think closing off, either economically or for other reasons, closing off platforms, I think, is something that's very challenging and it continues to be a challenge, particularly for a lot of people that I believe are innovators on this call. So one of the things that's passionate for me is what Cole and I have been talking about for many, many years, and he wears it as a badge of honor. I don't talk about my telco days, but it gives you a good perspective on what you weren't able to achieve then and you gravitated towards what we see as the new model. And I think we'll talk about it in more detail, and I'd love to hear your feedback around -- it's not just about space and power, right? And I think that's something that I think all of us would agree to, and it's more around the interconnection and how the culmination of all of those elements come to market is something that's meaningful to us, and we want to continually refine with this group. But I really see Edgecon as being a catalyst to drive innovation in a lot of the proof-of-concept markets that we're talking about today and that we've launched, and we welcome your feedback. We welcome you your workloads. But I really appreciate the opportunity, Cole. And thank you very much.

Yes. Thanks, Chris. And you should wear it as a badge of honor. Not many people can make the leap or do make the leap. But -- and not to throw Chris under the bus, though I try to do that whenever possible, he balks at Edgecon because he's in Hawaii every single year when we sort of run one of these events in Aspen. But to kick us off here, Chris, the first, and for those that have been keeping up with us and sort of our events, Aspen is one of those events that's turned into a summit for us. But Chris, you were invited to the first Edgecon, the very first one in Aspen. And I was back in, if I remember correctly, 2016. What do you think has changed for the Edge between 2016 and today?

Yes. No, I think it's a great question. I think there's been a lot of, I would say, hype on one side, right, where I think, in 2016, there was a lot of, oh, I always joke, and Cole and I have talked to a lot of media outlets with the fact that there was cloud washing, then there was Edge washing. And I think there's a lot of hype around what it could be. And I think what's changed is now there is more discrete workloads that are depicting where you need to place infrastructure and how that needs to be interconnected. So said differently, it's becoming real. And that's why we, as digital really in working with Vapor and just looking at the landscape, we're really wanting to put some proof-of-concept or sandboxes, if you will, out in the market to allow for that innovation to happen. So I would tell you, living in Boulder, love Aspen, would love to have attended. But I would tell you, the finer points of the work yield requirements that Cole and I have talked about this many, many times, the workload is going to depict what is required. And so we view ourselves as foundational in providing the capabilities and removing as much complexity as possible. But really, I start to see these workloads coming in that will further refine exactly what we have to deliver from infrastructure and services, but also be a catalyst to allow us to expand it in a material fashion because I think that's the last pieces. What's nice about the hype is that there's dollars that can be allocated to this space. I do think there's going to be a lot of wasted dollars in this space. So learning from your customers and allowing them to innovate rapidly on top of it is the last element that, I think, would say -- I would say is, here today, from a couple of years ago, is that the dollars are starting to be focused, and there's some bets that I think can be materially made in this space to really allow what I feel is going to be the next generation of the Internet, if you will, and really interconnecting out to some of these new spectrums and some of these new locations.

That's insightful, Chris. I'm -- for the Drew Carey fans here, there was a show in the late '90s called Whose Line Is It Anyway? Who's Edge is it anyway, Chris? Like we talk a lot about the edge. But like -- and some of us are really opinionated about it. Some of us are very curious about this multifaceted sort of 3-dimensional Edge. And what common themes apply no matter what vertical we're thinking about? So if you're retail, if you're life sciences, if you're an industrial agriculture, like give us -- where is the Edge and whose Edge is it?

Right. No, I think that's a great question. And it's nice to be a part of this group because you can really drill into the nuances of it. And I think what's important to us at the highest level -- and I would say people change companies, but friends are forever, and I love to give Cole a hard time. And we got asked the question this week around is the Edge a container in a field. And I'm like, "Yes. That's exactly what Cole built. It's a container in a field, and that's what the Edge is." That's what it's not. But what I love is in the highest sense of the term, one of the things we've talked about is core to Edge, right? And what that means to customers is they can build multi-tiered applications that have workloads that run in strategic locations across a gradient from a core data center for regional data center to an Edge data center. And the use case depicts where you place that infrastructure. And I think one of the interesting things to me and it's kind of something I even have it right here. Like I've been working with these Coral accelerators. This is a -- the TensorFlow. This is the device Edge, right? So these little -- the Edge is in my basement doing scrubbing on my videos, right? Like to get rid of false positives. But the Edge is really, again, back to the workload, and it's depicted by the workload. But what I think as an infrastructure provider to the broader community, and this is what brought Vapor and Digital Realty together and colonized kind of original discussions is you need to be able to provide that gradient, right? And I think this is where we all get stuck is you might put a blinder or a silo on one element of the infrastructure that has to be delivered to meet that one workload requirement. That's not what we're really talking about here is that the pervasive nature of providing an environment where you can deploy AI either way at the device Edge in a Vapor Chamber Edge or a local data center, even the core data center, there's a place for all of that, and there's workloads that require that, right? And I think every vertical out there, I think, is starting to open up to the fact that the infrastructure has to be pervasive enough to support their needs, and it hasn't been delivered today. I will say that. So this is somewhat new, and we welcome any kind of feedback, and we'd love for you to participate in our proof of concepts, and I haven't said it yet, but Dallas, Atlanta and Chicago, these are markets that -- where we really link the 2 together. And I think my last point with what the Edge means to us is there's a customer lens that I think you also have to look at that if you think the only one element is required to make it all successful, you will miss out on what I feel is a huge opportunity. So we service a lot of major hyperscalers, right, where we have the heritage within digital to do multiple types of data center environments, where we built a 100-megawatt facility, right? It's 100-megawatt plus single building in Ashburn, Virginia. But then we also have the heritage and expertise to do other more regionalized, localized facilities in all of those being heavily interconnected. But what we see is the supply chain from these hyperscalers and some of these workload and their infrastructure to be exposed and extended out towards that edge. And this is where we look at vapor as, they also have a similar viewer. It's not just about fleet and power. It's not just about that VIM and here's your multitenant shared rack and it's really cool, and like it's more about the software and the space and power and the network capabilities because it gives you that full resiliency within it. And that's when I talk about pervasiveness, going and having the ability to look at that entire supply chain and pick and choose where your environment is to where Vapor is -- it stood off a lot of its chambers and interconnected them where you can get resiliency within the ring because I think these Edge environments require a new type of environment where you can get a better return out of the capital, where you can run them hot, hot. I also think it's an important element that you start to work with some of these new spectrums coming in the market, right? It's not -- everybody talks about 5G, but there's a lot of other spectrums that the meandering of a multi-tenant environment deep in the metro with these spectrums, that's a sandbox that's not been created yet. But it is extremely valuable to innovators such as ourselves because just think about what you could do with that in the low latency, in the reach, I mean, that is what drives my passion and Cole's passion of what we've been talking about for some time and the further culmination of some of the press that we've done this week and -- has been launched today, but that's how I view the Edge.

So you said 2 things there that I want to kind of dig in on a little bit. You talked about infrastructure. And the infrastructure at the core is really similar to the infrastructure at the Edge. Of course, you talked about the network. And I do remember the reference to the lonely island, which coincidentally is also a comedy band. Raise your hand, if you have heard, "I'm on a boat, come on." It seems to me that a data center in the middle of a kind of a lonely island, and it's sort of a comedy in the Shakespearian sense of the word. And it can be a great disaster recovery location. But how -- Chris, how should this group be thinking about the network? And the dirty little secret in the telco world, of course, is it takes wires to offer wireless, but it also takes wires to offer wirelines. That's slightly more obvious. But talk to us about the -- what's important about interconnection and the network when you're talking about core to Edge or Edge to core.

And I'll talk about it kind of top down, right, where in the most simplistic fashion, we have a lot of users that like I've been I've been so happy to attend all of these like NANOG, GPS, like you name acronym, I've gone to the meeting, and there's such a subculture to them, which I love because you learn so much, right? Like in the randomness of some of these meetings, really, what I hear on the other side of the core, when they hear about GPS they're like, "What is that?" Like, "I have no clue on what is peering," and there's a bit of a fomo where everybody -- like I talked to more industry CIOs, CTOs, who are like, "Oh, I think we need to start peering." I'm like, "Maybe," right? Like they just want in on it. But removing complexity, that's like one of the biggest things that we look at, which is a part of this partnership in interlinking platform digital and our interconnection capabilities, particularly around the service exchange and our comprehensive interconnection portfolio in with Vapor, Kinetic Exchange and the tech. Removing complexity so that we've done the heavy lifting, and we've done the wireline interconnection piece in these markets to really remove that complexity so that customers can get a better visibility of that broader gradient or that broader spectrum to implement their infrastructure against. And I would also harp on the fact that taking one level deeper, interconnection is a game changer. It'll either help you or hurt you, right? And what I mean by that is the network elements of what we look at today at Digital Realty and some of the partnerships we formed with Megaport and some of these other providers, it's all about really equipping our customers in the infrastructure with more value. And customers get a lot of value, not only out of the space and power, but they get, I think, equally, if not more value out of the interconnection and who they can connect to and how they can connect to these critical destinations. And we see the edge as a critical destination be it a 5G, CRAM, be it CBR -- whatever it may be, the Edge is out there. But also a lot of these workloads require interconnection back to the core. And so drilling into that network, we've spent a lot of time at the core, where in certain markets, right-of-ways are a nightmare, right? And some of you are probably accustomed to this where we fight with municipalities and try to put down both conduit to critical destinations and run a very open platform that empowers our customers to communicate through our infrastructure into broader destinations, very similar, just to tie it back to the Edge and where I think you're going with this, Cole, is interconnection, in the way the tech has allowed a re-architecture to be crafted so that you can get resiliency and failover, it is still ultra-low latency with some of the new spectrums coming to market and being able to directly equip small cell, that is where I feel a material portion of the future is going to require that to be able to launch their workloads and their applications in a uniform fashion as well. But network is something that's near and dear to my heart. I mean I spent more years than I'd care to admit, looking at not only the products and how we deliver our products, but not like what -- this is one where I don't just talk to customers and say, "Hey, what do you want?" You really look at what the unknowns are because Cole and I have talked about this in a couple of occasions where if you ask a customer what they wanted way back when, faster horse, whatever analogy you want to use, the network, because it's not here today and people have different views of it, they don't know what to ask for. They don't know that there's a seamless environment where you can pick up a core all the way to Edge, and there's this metro ring for failover and interconnectivity. So at the end of the day, I think it's an equally, if not more important, element as we stand up the infrastructure portions and how you get interconnected and how you let that interconnection flow.

That's awesome. If I synthesize and restate, and I don't want to put words in your mouth, but what I sort of heard is if you care about the nuts and bolts of the Internet, like a lot of the people at this event do, the killer app for the Edge, in our world, can be boiled down to 3 things; physics, the easy button and economics, is that right?

Yes. Yes. Maybe you're spot on. You're spot on.

Okay.

And I would -- there's something else. I mean when you talk about the physics and stuff, there's a lot of work that we've put into data graph, right? And I think you had said this the other day, Cole, which I think it's important for this group because they'll get the next nuance to it, where a lot of our core facilities, data lakes, data oceans, lots of data resides there. There's a benefit to that. A lot of ecosystems are created around that. But I think the other element that the network enables is data velocity, right? The speed at which you can interconnect. And if there's economical barrier or some type of technical barrier, that's something that the combined Vapor-Digital Realty are constantly looking at how to overcome those, but I would put that finer point into it as well that the data velocity element of the network is another critical piece.

And it's funny, thanks for that. We're going to wrap up here. I'm going to come back to killer apps because we like talking about those things. But it's funny, as I look across the room with people that I've worked with in the past, the Internet guys, we tend to measure things in millisecond. I'm actually looking at Michael, starting to call you out from T Mobile, but telcos have to think about things in terms of microseconds. 75 microsecond front haul is an absolute requirement for a lot of remote radio head functions, et cetera. So even the data velocity for the -- for one first part of your third-party service might look very different than say, the front haul, mid-haul or backhaul for another first or third party service. So I guess, thanks, Chris. I know we're already a few minutes over. I would, though, like to -- and something Chris and I chatted about, the only thing we chatted about before this talk was how do we continue the dialogue today with everyone on the call? And one of the things that Vapor is always excited to hear about, and I'm sure now, given the news that just launched this morning with PlatformDIGITAL service exchange and the Kinetic Edge exchange being launched in additional cities, more killer app ideas. Like what is interesting to you guys in terms of an Edge to core or core to Edge architecture. Is it backhaul elimination? Is it something -- is it robotic surgery? Is it autonomous robots? Like we want to hear from you guys about how to build these reference architectures. And I think Satya might want to hear about this too because he's got a lot of PhD students that are thinking through their thesis, and there's no shortage of talent over at Carnegie Mellon thinking through these things. So I'm very curious about that. I know a lot of other people on this call are very curious about that. So what a cool way to kick off of Edgecon. Really excited to break out into some of these breakout rooms with you guys and I, for one, will be sticking around to listen to Satya and the rest of the call today. So [ Brianna ], I guess, if you're going to throw us into breakout rooms now. Is that right?

Yes, absolutely. And for anyone that joined, I am putting a quick note in the chat. I know a few of you all joined late. So we're just going to force you all into some breakout rooms, talk about that prompt, and we'll bring you back in about 5 minutes.

[ Brianna ], I just want to say, I was in a lonely island in the last breakout room. I was in breakout room 3 by myself.

I'm not going to call the people out who -- there were a few people. If you do end up in a room with -- by yourself, please come back to the main room, and I'll reassign you to a room. Don't want anyone sitting on a lonely island, unless you have a jinx and you're...

We don't either, and that was awesome. We don't either, [ Brianna ]. Thank you. Thanks, Chris.

Thank you. Appreciate it.