Ciena Corporation (CIEN) Earnings Call Transcript & Summary

Great. Welcome everybody. This is Meta Marshall. I lead up the networking research here at Morgan Stanley. I'm going to read a short disclosure, and then we will jump into our discussion with Stephen Alexander, the CTO of Ciena. So please note that this webcast is for Morgan Stanley clients and appropriate Morgan Stanley employees only. This webcast is not for members of the press. If you are a member of the press, please disconnect and reach out separately. For important disclosures, please see the Morgan Stanley research disclosures website at morganstanley.com/researchdisclosures. If you have any questions, please reach out to your Morgan Stanley sales representative. And with that, I will jump into a conversation with Stephen Alexander, CTO of Ciena.

So Steve maybe to start, what stresses has COVID brought to your service provider customers, both in terms of demands on the network and the ability to serve those customers?

Well, so Meta, I think, and I'm pretty sure people have heard this, the whole industry saw almost a year's worth of growth in terms of capacity demand in something like a few weeks to a month, right? As everything went to kind of this distributed digital environment, the demand moved from downtown areas out into the suburbs. Capacity demands went up. The peaks of traffic changed, it became much earlier. It wasn't people going home and watching streaming video. It was folks doing like what we're doing here, getting on a video conference and getting on Zoom, engaging in the normal business meetings, but now overall on a distributed basis. So the way I've always described it is take that years' worth of growth and just apply it into the network in a couple of weeks to a month's time.

Great. And how were they able -- clearly, there were kind of some installation challenges. And so how are they able to accommodate that just given some of the challenges of accessing sites or making changes to the network?

Well, it varies. Folks -- and again, there's a broad spectrum of service providers out there. Those that had -- kind of already had in place upgrades for capacity connectivity did relatively well. Those that had already gone to the point of saying, I'm not building a number of parallel networks. If you go back historically, a lot of the telecom infrastructure grew up initially as based on the old phone system. And then on top of that, we added digital capabilities. Maybe we've built an enterprise network again in parallel to that, and on top of that, we might have built a wireless network. So there are some networks out there that literally were built as almost 3, 4, 5 parallel networks. Many of the service providers grew through acquisitions, different combinations of things. And so there's a -- there's quite a variety of approaches. If they had already started down the path of modernizing their core, their metro cores, their long distance cores, their inner city cores, they tended to fare a little bit better. One trend that's been out there for several years now but COVID has certainly accelerated is the move towards kind of a universal access platform, where you just bring as much capacity as you need right up to get into the network. And then you assign how much of that capacity goes to residential, how much goes to enterprise, how much goes to wireless at that location. You don't do it by building parallel networks and pulling them back into core, you bring all your capacity out and then you sign with the edge. Once you make capacity fungible like that where I can start to exchange it between service types, you get a much more flexible, much more adaptive kind of a network. And that's been certainly a foundational philosophies of Ciena's for quite a while. Those customers that had kind of started down that path where they had made capacity more fungible tended to fare better.

Got it. That's helpful. So you spoke about kind of the stress that the service providers saw on their networks. How is Ciena able to accommodate those changes? And where were you kind of in a unique position to help those customers?

A couple of places. I think a lot has to do with the way we've constructed our supply chain. By design, we don't supply parts from China. Our final assembly is North American centric. We've taken, over the last decade, decade plus, a number of very deliberate steps to build what we believe is the best supply chain for telecom optical and packet optical in the world. And that fared -- I think, served us very well, served our customers very well in this pandemic. They were able to get capacity. To your point, there were the site access issues. If you were going into a location where you had quarantine requirements, there were some that -- just to get into the country, if you didn't have in-country assets, you had a 1-week, 2-week kind of quarantine period that just slowed things down. But the way we've built the supply chain, the fact that so much of our -- the infrastructure that we provide is software controlled now, you can remotely configure it, you can change it. Much of the basic photonics that we deploy is programmable right down to the rate on the line, right the way you modulate the light. Being able to change that program and redirect it all made the networks more flexible, more adaptive, right? And getting it to be adaptive is going to be critical for the future. I mean we've learned that, right? But if anything, we've all had to adapt to new ways of working in the sense that the networks had to adapting new topologies and traffic models, just everything kind of changed very quickly. But fortunately, if, again, you had kind of thought about that in advance and said I wanted things to be more programmable, more software defined, more intelligent, generally speaking, those networks did better in this kind of a very stressful environment.

Got it. That's helpful. And just as your conversations with customers, how are -- kind of like how are service providers trying to determine whether this is a permanent change? And they'll just have to accommodate more distributed traffic going forward. Or how will they accommodate more real-time video going forward? Just like how are your discussions going about permanent versus temporariness of this kind of COVID situation?

Well, I think most of the service providers news that, especially with the advent of 5G and the kind of services, the augmented reality, the virtual reality, the immersive sorts of things that are going to develop over, let's say, the next decade or so, they were eventually going to get there anyway, right? The kind of final state was identified in terms of lots of capacity for the edge, lots of dynamic nature in the network, lots of these immersive, interactive kind of applications were all coming. COVID accelerated all that by quite a bit. And so I think -- I haven't run into a service provider who's looking at this and saying, oh, this is all going to back to the way it was. I think what they've looked at it and they've said, these are kind of permanent changes in a lot of ways. I think that's true in general for our working style. A lot of things have gone digital, more than any -- ever before, where it's more distributed than ever before. I'd also say things have to be more deliberate than ever before, right? You have to plan your interactions in such ways that you never had to before. And I think, in general, the service providers are looking at this and saying, yes, these are going to be some fundamental, permanent changes in how we have to think about the infrastructure. The other thing that's gone on, by the way, is what we've been doing on the technological front, pushing more and more capacity out. We have this kind of soundbite that we like to use a 400-gig everywhere. But I think it's important to just go back and reflect, right? 20-something years ago when Ciena started, 40-gig worth of capacity was a full bay's worth of equipment, right? And so you imagine a full telecom bay, 6-, 7-feet high, 23 inches across, I mean big, right, for 40-gig. Next year, we've already got prototypes of this, you're going to put 400-gig and something that fits in the palm of your hand. That's a phenomenal reduction in the size, space, power consumption. And what that all translates into, of course, is that means the photonics and optics and the technologies that we've pioneered can go that many more places. In fact, we would say can go everywhere, right? And the fact that you can bring so much capacity out to the edge of the network now means you can think about having these universal access platforms in much more dynamic, adaptive network architectures than we could in the past where we were just really aggregating lots of thin traffic flows coming in, you need massive amounts of aggregation before you got into the network. If I can start to originate and terminate 400-gig silos right into the edge, I have a different architecture I can build.

Got it. So maybe continuing on that theme, clearly, bringing more capacity to the edge is important. But how does the network topology change with more kind of distributed traffic as they're more, I think, in like more packet networking, more metro? Just like how do we think of how the network architecture changing?

Well, so I think you're going to see a couple of major vectors pushing, right? One, to your point, more packet optical. So the same network element that can create and terminate the light will also handle the packet flows, right? That's a level of integration and convergence that we've been driving for several years now, that's going to continue. The other one that I think you're going to see out there is the cloud, the edge of the cloud moves closer to the user, right? We have this term, I'd like to call it mean time to cloud, right, to try to kind of identify where the edge is because there's so many different definitions of what the edge of the network, the cloud, all these things look like. I just talk about it and something you can't change, which is a physics, just how quickly do you want to have that cloud experience? And you find that to get the numbers to line up for things on the 5G side, for the AR/VR world, for tactile Internet, you're going to need numbers no longer than 10 milliseconds for things that are really interactive. And you start to just work through the physics, so okay, if that's the requirement, and I've got an area interface I've got to get through on the wireless side. I've got some electronics I've got to get through. I've got to get to that cloud. Where does the cloud edge have to be? And you're going to start to see now edge compute capabilities showing up no further than maybe a few hundred kilometers away from you, in many cases, even shorter, tens of kilometers away from you for certain kinds of applications. And that's going to be your first cloud experience. And so the other architectural change that's out there is if you take the telecom infrastructure, the cloud is basically building right over on the top of it, right? And there's the carrier infrastructure here, the cloud infrastructure up here. In some cases, they're the same. They're built and operated by the same person. But in many cases, there's multiple carriers and there's multiple clouds, and it's this multi-cloud environment that's emerging, which is really going to be next frontier. In terms of the functionality out of the edge, they've got to be built with the understanding that it is inherently going to be a multi-cloud environment.

And just where do you think service providers are on that journey? They may have -- you kind of spoke to, there's a wide variety. But just on aggregate of how they're thinking about or how they're evolving towards some of these new changes that they need to make?

Well, as we look across the globe, there is tremendous variety, as you would expect. In certain areas, certain countries, the service providers want to be the same cloud operators. In others, there's a clear separation. And I think the model that probably is adopted first in a lot of locations is I take a portion of a switching center or a central office, I'll cage off a portion of it. And the cloud operators will come in and they'll rent space from you. I'll give them space, power, connectivity that they'll put their networking equipment there inside of the carrier infrastructure, if you will, right? Get traffic off the carrier infrastructure and add onto the cloud as quickly as possible. I think that's going to become a fairly common model as they learn how do these things all work and what does it mean when I've got a central office, it's a service provider, and I've got multiple bays from different cloud operators and how all that plays out is the work in front of us in some sense. That's this kind of multi-cloud environment.

Got it. And then just in terms of -- you've talked about software kind of already a couple of times. You clearly have the software business with Blue Planet helping service providers automate their network. How does this become a growing part of the discussion and particularly as kind of traffic pattern become more fluent?

Well, that's exactly why automation is so important. The example that I keep going back to that this industry did accomplish, right, is the smartphone. If you go back and you look at the old feature phone, we each have our feature phone. We flip it up. And some combination of keys, we'd read our instruction manual, we just figure out how to put each other on speed dial, probably takes 10 or 15 minutes to do that. There's so much software intelligence built into that smartphone that you can give it to a child and they're going to figure out how to use it without an instruction manual. You don't have instruction for it, you just kind of figure it out. That's because there's so much software. The same thing has to happen in the networking space. The way that, especially now with COVID, where sending someone to a site involves -- may involve quarantine procedures, it may involve testing, it involve all sorts of things that just slow down the work. Having the ability to do things in an automated fashion is just far more important than it's ever been before. I think the question that a lot of the service providers have is look, how do I get started? What's the foundational moves I have to make to assure myself as I continue to automate more, I don't go off the rails. I think that's a conversation that we have a lot with folks because it's like, I know I have to automate. I don't want to be doing truck rolls for all these things anymore. COVID's proven that to me. What do I do, right? And that's a great conversation to have because you start with an understanding of what are the basic processes that you've got in if you're a network operator. It's usually planning, planning to build. Where do I have to put infrastructure facilities to meet my customers? How do I grow? How do I grow my market share, right? There's kind of the -- from the time I take an order to the time I can provide service. And then there's from the time service is established, something happens, it's kind of the break fix. But those are the 3 basic things they've got to do. Where do I put infrastructure? When I get an order, how do I turn up a service? And when something goes wrong, how do I fix it? And when you really distill down through all that, the thing they have to start with is knowing what have they got, what's their configuration, what's their inventory position, what's there, right? And there is the physical devices, the boxes, the virtual devices, the customers, the connections, the flows, all that has to be known really well, very accurately. It can't be a bunch of manual reconciliations between technicians' notebooks to figure this out. It's got to be known in a way that you can automate it. But if you do that, if you take the time upfront to do that work, then you can really automate your infrastructure really, really well, and we've got proof points around that now. So we've got some customers who have started down that path. They're now at the point where they can say, yes, we really see the benefits of this. That's the proof point that you can take and go to other customers and say, hey, look what these folks accomplished. Hence, paying attention to the fundamentals, knowing what they had, look what they were able to accomplish from an automation point of view, that's very powerful.

Got it. I mean, another way that people will think about simplifying a network or people talk about supplying the network is the use of pluggables. Just how does that come into the equation as far as where you can be helpful where customers are looking for kind of changes to architecture?

Well, so it's absolutely my comment very early on around 400-gig in the palm of your hand, that is a pluggable by design, right? So we looked at it and they said, Ciena was privileged to be the pioneers of 2 technologies, dense wavelength division multiplexing and coherent optical detection. You put those things together and you've got all the makings we need to make these very high-performance pluggables. And that's exactly how you get photonics and optics to go everywhere is you make it something you can use in multiple places. And so that's one of the things we have pushed and contributed to. We've been a driving force behind the adoption of the standards that drove the pluggables, all those sort of things, right? So our view is, you put out at the edge of the network these universal access capabilities. And plug by plug, you can define whether it can go to a wireless, whether it goes off to enterprise, whether it goes to consumer, residential, whatever you want, and then you can allocate capacity amongst it, right? So we're actually fans of the pluggable piece of it, right? You have to be careful when you get to the higher-performance portions of it because you're back into the physical propagation and it's kind of more of a natural analog world. But out at the edge of the network, we're big fans of pluggable. We use them today.