The pandemic has been an interesting crucible, showing us how well we can work together as a society but also exposing the digital divide between those who even have that opportunity and those who don’t and never have. With us today to offer his thoughts on how today’s world is doing when it comes to leveraging technology for all is longtime industry veteran Steve Alexander, SVP and CTO of Ciena. Steve has been at the forefront of many of the technological developments that have shaped the modern world.
TR: What is your background? How did you get started in optical technology and what was your journey like to your current role?
SA: I have been at Ciena since 1994 when the company started. But before that I was a research engineer for MIT Lincoln Laboratory, and I was always interested in communications. I think it brings people together. When I was a teenager, I saw a helium neon laser and learned that you could modulate it and send information over a distance with light, and I just thought that was so cool. The last project I worked on at MIT Lincoln was called the whole optical network, which eventually brought me to Ciena where we introduced DWDM and the control plane. We survived the telecom nuclear winter and eventually acquired a portion of Nortel, which we have managed to leverage very successfully. I like to describe Ciena as a story of abundance: abundance of capacity and abundance of conviction. We light the fibers, we bring the technology, and we enable sheer capacity out to the edge of the network. Consider the fact that we’ve been able to do so much over Zoom etc. now. Yes, there were a few issues at times, but the fact that so many people can successfully work from home and keep the digital economy going is a tribute to the kind of network technology we brought to the marketplace.
TR: Yet challenges still remain. What is the digital divide?
SA: I think people still talk about the digital divide as whether you have internet access or not. But I really think there’s three components to closing the digital divide. Number one is, ‘do you have Internet access or not?’. Number two is whether users have devices to talk to the internet. And number three is whether they have the skillset to use those devices. You have to solve all three of these basic things if you’re really going to solve the digital divide. We could go in our lab in Hanover Maryland years back and show a terabit connection on a fiber. But drive 20 minutes away to Annapolis and it was DSL and cable modems, and cross the bridge to the eastern shore and most people didn’t have any internet access unless they were directly under a cell tower. That rural problem has only recently started to improve. But on the inner-city side, such as in Baltimore, you can see today what happens when the kids don’t have access. They just can’t experience the same things that the more suburban kids can. I think it manifests itself in multiple areas, built into the fabric of society.
TR: Will recent government-led efforts, such as those in the recent stimulus and infrastructure bills, do much to address the digital divide?
SA: I’m not familiar with all the pieces, but I think it will help. Certainly, government efforts will help folks put up the antenna, light the fiber, and just get the connectivity out there. I don’t know how much of it will go towards devices or how much will go to training and all those other pieces. But I think it’s a good start. When people think about government investment, they go back to the Rural Electrification Act. What was the business case to put electricity out to all the farms? Well, there probably wasn’t a very good one, but people knew it was a good thing to do. I think you’ve got something of a parallel story here. There may not always be a great business case to run connections out to the farthest extremes of the country. But if there’s people there and if you want them to be able to partake in this digital world that we’re creating, where we can work from anywhere, learn from anywhere, get health care from anywhere, you’ve got to get them connected. You have to put these things in place knowing that over time things will get better and applications will come along; the problems will get solved.
TR: Are there ways Ciena is taking direct action to address these other components of the digital divide?
SA: Ciena is investing $10m over the next 5 years to help underserved students. We have programs with different partners, and we are trying to make it be more employee-led so it’s different in each community. The folks that are local know the most about what’s really needed in their area. For example, we are working with Verizon in Baltimore and San Jose, where we have picked two underserved schools, equipped them with devices, and have set up local employees to do hands-on training and development to get the students educated on the technology. The digital divide is a problem that no one company can solve. But if every company does what they can with the knowledge they have, we can make a huge difference.
TR: What about beyond the US marketplace?
SA: Ciena may be a North-American-centric, US-founded company, but we have a global customer base. The issues that we face here in the US have some similarities, but are also different from some place elsewhere in the world, especially places like India, Africa, portions of Southeast Asia, etc. There are different generations of infrastructure, different levels of technological debt, and different political scenarios. There are some national governments that know it would be better to have good infrastructure in place to get all their people on digital engagement. But not every country thinks that way. Ciena has broad exposure to what is happening because we sell to carriers and webscalers globally. Many of our customers’ addressable market isn’t just the US, it’s the whole world: anybody or anything who is connected at any given time. They are all in favor of anything that helps with that expands their addressable market. That’s one of the reasons you see so many different initiatives, whether it’s fiber or terrestrial wireless or satellite-based wireless.
TR: Where are we as a society when it comes to solving these problems?
SA: As a society it’s absolutely a mixed bag. There are some countries where it will still be more important to get running water, and there’s going to be others where they are well along a residential fiber build-out program. Only for the latter can you even debate whether they done a good job of getting devices into everybody’s hands and trained kids them how to use them. Do we have really, really good distributed learning apps? No, I don’t think we do. I’d say the business side of it is probably going to be in the best shape just because there’s so much money involved in getting folks to be business-effective in the world during COVID. We still don’t have all the desired health care centers of a low-cost point for people really do telemedicine well. We are making progress though.
TR: In what ways has COVID driven technological change when it comes to collaboration?
SA: I think there’s more awareness than ever because of the COVID experience. There are an awful lot of people that only got comfortable with a virtual environment because of COVID. It accelerated so many things in terms of the adoption of the digital environment of running meetings differently. I remember, with our homeowner’s association when we first couldn’t actually get together at the community clubhouse. The first meeting or two, it was tough, people just weren’t used to that environment. But now, the quality of the meeting is even better if for no other reason than that on the screen, everybody is the same size square. We no longer have the three or four people at the front of the room who are running the meeting and everybody else is the audience. Everybody talks. Everybody gets a voice. I think those changes toward a collaboration environment are just the beginnings of what is going to be coming next. I know there are folks who are working on the apps to do an even better job.
TR: Is the technology ready to support the rapid shift closer to the edge and to eyeballs?
SA: The move from 3G to 4G LTE, which brought fiber connections to towers, bought gigabit capacity to the network edge. With the most recent network improvements, we can do 400G in devices that fit in the palm of your hand. That means that we can start to move these high-capacity technologies anywhere in the network. I don’t think that has fully soaked through people’s understanding yet, because we’re so used to thinking high capacity must be big iron. Well, not necessarily anymore. When I talk to outside plant folks, the ones at the pointy end of the stick trying to get things done, I tell them the fiber they are putting in the ground could be carrying terabits per second in the future. You may choose to light it with gigabits today, just because that’s all that your customers need or that’s all that’s available. We used to think of the network as very skinny out at the edge, with lots of thin connections that you would aggregate together to create capacity. There is architectural change coming which will be driven by new business consumer applications. Think about the next generation of applications where you have an immersive environment. We might all be holograms interacting as if we are physically there. To make that happen will require a lot of capacity, a lot of computational resources, and very low latency. That means you better have at least some of your cloud workloads relatively close to you, and that means you’re pushing big connections to the edge, putting compute capabilities out closer to the end-users, and building thicker connections out to your final customers.
TR: Where is Ciena putting its efforts on the technological development front right now?
SA: Ciena develop equipment and push the bounds of capacity within the network. We were first to do 100G, 400G, and 800G, and we’re just going to continue to push to just the raw capacity per wavelength point of view. We’ve incorporated more and more packet IP routing and switching capabilities for the simple reason that on a chip that’s of a given size, given the continuation of Moore’s Law, you can just do more. Another area we have been focusing on is driving the adoption of universal access technique. The action for the next decade is out at the edge of the network. That’s where all the services are really going to get defined now in the industry. Everyone will have different requirements of the network, but the most common will be connectivity, capacity, and latency. People, in general, don’t understand the implications of latency in the network. We’re just not used to it in terms of real-time interaction. We’re very used to streaming now, and if you’re off by half a second who cares. But in a collaborative environment you might be looking at graph #3, someone else is on graph #5, and another is on graph #7, and we can almost in real time update each other’s work. But to make all that work, you have got to get latency under control. If I’m building out 5G infrastructure for the “Factory of the Future” in which the workplace stays put and robots move around it, your latencies have got to be milliseconds or even hundreds of microseconds.
TR: The low latency world has been dominated by high frequency trading, which focused on specific route optimization problems with algorithms on both ends. Where else is the latency, and how do we approach managing it for more complex usage?
SA: The high-frequency trading guys were always going to be there. But when I say you’re going to have to get latency under control, what I mean is you’re going to take an infrastructure and you have to imagine a hierarchy of clouds. The edge means a different thing to every person you talk to, but if you ask what latency requirements define that edge then all of a sudden you can get back into some real engineering kind of calculation. You are going to have three data centers to pick from. One that’s going to give you latency that’s single-digit millisecond, which will probably be in your state or even in your city or town similar to a central office. You’re going to have another data center with latency that might be tens of milliseconds, such as the typical hyper-scale data centers you see scattered around the country. And then you’re going to have the truly gigantic hyper-scale data centers that might be 100 milliseconds from you, even across an ocean. Your experience will be governed by where those workloads run.
TR: What areas do you see service providers and vendors putting most of their development effort going forward?
SA: When I talk with folks, there are 5 topics that generally come up, summarized by “faster, closer, smarter, safer, and greener”. Everyone always wants more capacity of course. But they also want the network to bring the cloud closer, and they want their whole infrastructure to get smarter. The smart piece of it really speaks to service velocity. Think about the experience that we had when COVID hit and we had to work from home. We could download Zoom or Webex or GoToMeeting and in five minutes get it up and running. By contrast if I needed to order an Internet service or change my Internet service, it would be days to weeks to months. That’s a very large gap between the velocity at which the cloud operates to provide you a new service and the velocity at which a service provider can operate. Closing that gap requires intelligence. How do we just make the whole network more real-time? Of course, safer is about network security, and greener refers to the whole power consumption problem and climate change.
TR: Thank you for talking with Telecom Ramblings!
If you haven't already, please take our Reader Survey! Just 3 questions to help us better understand who is reading Telecom Ramblings so we can serve you better!Categories: Industry Spotlight · Low Latency · Telecom Equipment · Wireless