Part 1 of 2 - Future of Tactical Communications & Computing at the Edge

Welcome to “HSDF THE PODCAST,” a collection of policy discussions on government technology and homeland security brought to you by the Homeland Security and Defense Forum. 

In this first of a two part series, hear how DHS is embracing the cloud, the strategies they're employing to migrate systems, and their forward-looking plans for edge computing.

Featuring:

• Craig Basham, Deputy Executive Director, IT Operations, Office of the Chief Information Officer, Department of Homeland Security
• David Earle, Director, End Users Services Division, Transportation Security Administration
• Charlie Armstrong, Chief Information Officer, Federal Emergency Management Agency
• Jim Westdorp, Chief Technologist, Ciena Government Solutions, Inc. 
• Dr. Dan Gerstein, Former Acting and Deputy Under Secretary, Science & Technology Directorate, DHS (moderator)

This discussion took place at the HSDF Technology Innovation in Government Symposium on February 22nd, 2024. 

Follow HSDF THE PODCAST and never miss latest insider talk on government technology, innovation, and security. Visit the HSDF YouTube channel to view hours of insightful policy discussion. For more information about the Homeland Security & Defense Forum (HSDF), visit hsdf.org.

Transcript

Dr. Dan Gerstein: 0:01

Well, good afternoon. I think we've got a terrific panel for you today. What I like about it is that it covers at the department level. Of course, we're going to talk about some things and then, also, coming down, we're talking to a couple of different components, which will be great, and we're going to have an industry rep to sort of set the tone. So I'm very excited about this discussion. You know, I'll just say I won't give you a pitch for personnel like Luke did, because I don't want any competition when I go talk to Luke after about getting back into the department. I mean, he made a real pitch for that.

Dr. Dan Gerstein: 0:38

So, a couple things starting us off. We've got a number of different questions that we want to have laid out. We'll also have about 10 minutes for questions and answers from the audience, so that'll be a good opportunity for you to raise any questions that you would like to hear about. So the first thing that I'd like to sort of lay out there and ask our industry rep to help us out. First I'm going to pick on him and I'd like to have him give us a bit of a tutorial about what is this thing that we're calling tactical communications and computing at the edge. How should we be thinking about it? And I'll also prep by saying that some of the government folks on the panel are in the early stages, so there's a lot of work to be done, so I think this will be instructive for them as well. So, please, okay, yeah, sure.

Jim Westdorp: 1:31

Really happy to be here and talking to everybody about this subject and as Siena being a infrastructure company who puts together networks and computing infrastructure, something that's near and dear to my heart. So when we talk about tactical communications and edge compute, things like that, this has been an evolving capability. It used to be that comms were relatively slow, compute was more expensive and not particularly ubiquitous, and over the last couple of decades but of late it's really gotten to the point where we're able to push capabilities towards the edge that historically we've had to keep near the core or in the data centers. So what we're working on in industry is really kind of trying to take the network and take the communications infrastructure and take those limitations that used to exist and make those go away, so that the problems you have to solve are the mission problems and they're not bounded by your communications infrastructure or really whether you're at the edge of the network or closer to the core, and so the kinds of things that we're, as industry, bringing to market and that we're, you know, the different enterprises are now able to use, are things like, of course, when you're talking about the network itself, you got to get some of the high-speed networks you know as close as you can to the edge and those capacities have been really going through the roof. You know we're able to bring, as long as you can get fiber or even some of the higher speed radio technologies. You can get gigabits, even terabits of capacity. You know at least close to the edge of the network. And then when you're at the edge of the network you're talking about you know various forms of wireless. It could be CBRS radio, it could be SATCOM. Now some of the things that we're looking at is getting 5G technology, so things that are interoperable with the handsets that all of us carry around.

Jim Westdorp: 3:51

In the first place, getting those to the edge. And even if you're outside of traditional Verizon, at&t coverage, you can get mobile cell on wheels. You can bring that capability to the edge so that your users can use it with the mechanisms and vehicles that they're used to, and then still be able to get it back. And then again it used to be you had to backhaul a lot of information because you didn't have the compute necessary to do real-time computations. And even some of the previous panels talking about some of the AI a lot of the AI is still done back in the core of the network, but you're able to do this more and more at the edge so that you're able to get latencies down and more real-time.

Jim Westdorp: 4:39

So it's really an evolution and really a sea change in the ability to solve the problems that you have closer to the edge, closer to where the mission is, and not having to backhaul everything. The flip side, though, is when you do need to backhaul, we've got nearly ubiquitous bandwidth and compute deeper in the network to solve some really hard problems. So we're able to partition problems in ways that you really couldn't before and make it about solving the mission problems and less about. We can only solve it this way because we don't have the compute or network infrastructure. Now, those are pretty much there.

Dr. Dan Gerstein: 5:22

Super. I mean, that's a terrific laydown of the capabilities that are envisioned. What I'd like to do now, though, is ask the government representatives to talk a little bit about this idea of operational missions that they see on the edge, and you know, as we talked about earlier, you know some of these are aspirational, and so I'll ask the government reps, as they go through, to say you know, this is something we're hoping for in the future, whereas, you know, point out those operational issues or missions that are already ongoing. So let me begin with Craig. If I could All right.

Craig Basham: 5:59

So, like what you're saying, by the way, by pushing that out there. I think one of our big mission sets along the way has got to be getting that data out there, and it's got to be video, voice and data. One of the challenges we always have is the amount of data when back to headquarters, wherever, and actually be useful data and useful video. So one of the things we're really looking in, one of the challenges is how do you do that across 5G in a mobile setting? Yeah, we can wire it to a tower, we can wire it to a building, but when we need to be mobile with some of our partners, how do you get it out there and actually push it back and be solid data and useful data for operations?

Craig Basham: 6:50

So one of the things where we continue to work on with the components from the headquarters perspective is one, the operational mission, what is number one? One, the second is what is the data you need and when do you need it? So that's one of the big struggles that we've always had is we can come and feel the dreams perspective and we can build it and they will come, but a lot of times it's always too much, it's too costly, it's not secure. So we've got to keep pushing and working with components to do that whether that's video, whether that's voice or that's data and getting the type of data that they need out to the edge.

Dr. Dan Gerstein: 7:27

Great. So, Charlie, give us some perspective on FEMA. What's your goal for the?

Charlie Armstrong: 7:35

edge. I see Latwita out there. Maybe I should recruit her to help me here. You want to phone a friend? Yeah, phone a friend. Using the edge Exactly, using the edge Exactly.

Charlie Armstrong: 7:48

So you know, in typical response, I think that FEMA has a lot of experience at getting communications into. Not that there's anything normal about a disaster, but a hurricane in Florida. We know how to get comms on the ground. We know how to get deck MERS in there, get things stood up, get comms back up and running and start to reconstitute things.

Charlie Armstrong: 8:31

This year I think we saw a huge shift with Typhoon Maywar and I'm still learning the difference between a typhoon and a hurricane and with the fires in Maui, where we did not have the ability to get comms in there in a really rapid fashion, right. So you know, we had to quickly pivot and go to low earth orbit satellite through, in this case, starlink, and stand up some comms to be able to back end into our data center. We're in the process of migrating those systems out of our data centers and into cloud and we really believe that that's a key part of our journey towards really using the power behind Edge, because if we get those systems in cloud and we get them dispersed across the cloud community. That gives us access to a number of points of presence where we can then use edge computing to do some of that on the ground work, while the rest of the comms are getting stood back up and then back in that data into those systems so that it can later get uploaded and finish processing.

Charlie Armstrong: 9:54

But you know, I liken it to the work that I've seen going on in the Border Patrol today and I know I'm no longer at CBP, but you know they've really done a lot with edge in terms of trying to take care of trying to utilize what they call that dead time where they're transporting people to come to do some data collection, do some computing.

Charlie Armstrong: 10:17

We want to be able to do the same thing and whether that's, you know, administering flood insurance claims, whether that's flood insurance claims, whether that's individual assistance claims, whether that's any of the benefits that we give to survivors, and or just working with our whole SLTT community to share information, be able to understand what the operational lay of the land is and help standing and reconstituting back up things like communications, transportation, all those critical things. Edge, we believe in the future, is going to play a really key role in helping kind of take care of some of that space between you know, when we first get boots on the ground and the response to when we're actually starting to do recovery. So I think there's a great future there for us.

Dr. Dan Gerstein: 11:19

So let me just follow up on that, because you have a unique mission. I mean, you come in when there's been a disaster, when there's been some issue, and you're not necessarily falling in on a capability that's already there. So have you thought about how do you keep that edge vibrant in the midst of some sort of disaster? Is that a problem that you're dealing with?

Charlie Armstrong: 11:44

Well, honestly, dan, I don't think we're that far along on our thinking yet. Well, honestly, dan, I don't think we're that far along on our thinking yet. But what I have learned so far about edge is there's so many different ways to kind of crack that nut, whether that we can get to and kind of put up some initial systems capabilities and have that downlinked to whatever our handheld devices are, mobile devices. I think there's such a variety of those things coupled with things like Starlink and the emerging because you're seeing more and more of these low-Earth orbital companies starting to formulate Hoping that market's going to continue to kind of push along so we get some competition in there too. That, coupled with our traditional push-to-talk comps, I think really does help us move towards edge. And if you kind of look at where the traditional tactical communications is going, they are starting to look at low bandwidth kind of comms traffic going over that. So we could leverage some of that also to take care of some of our tech needs.

Dr. Dan Gerstein: 13:11

So let's hear from TSA David, what do you have to say?

David Earle: 13:16

Thanks for the opportunity to be here. Tsa our mission being out in the front lines of the airports securing the traveling public. There's a lot of data that sits at that checkpoint. Tsos transportation security officers need data to do their jobs. I'm sure you'd be surprised to hear that we can't give every person a piece of government equipment to get that data. So that's one of our struggles is getting that data to those frontline workers to do their mission and get there. So we're looking at some things to get the edge computing going. I focus more on the device side, so for me in my world it's getting those devices out that we can then use that data, process that data and then even on the other side, input that data that's coming from the checkpoint back to the people at headquarters or the coordination centers that need to make those decisions. I think there's a great future. We were talking earlier we're still in our infancy a little bit of edge computing. So I think there's a gold mine of stuff for TSA to leverage there Terrific.

Dr. Dan Gerstein: 14:20

So the next question I'd like to sort of pose to the group is to think about how far you are along in your transformation efforts and try to characterize it for the group, because I think industry would like to know where there are opportunities. You know at times you may need help with just thinking through the technology. Other times you may be thinking about the devices themselves or the networks or the ancillary equipment. So it would be useful if we could just sort of go down the line, starting with OCIO, and just hear a little bit about your thoughts.

Craig Basham: 14:59

Okay. So From a headquarters perspective, we are still trying to find that ability to do interoperability across all of the components and provide some type of data and solutions, for that will cover multiple things. So one of the areas that we're looking at right now is we have a pilot going on around the National Capital Region and the National Mall where we're looking to get where there's limited data or limited spectrum, there's limited availability of video, voice and data, because everybody needs it around that national capital area and the National Mall. So we're looking to work with partners on this to start really finding solutions where you can start doing network slicing, where you can start doing the slicing, we can start doing the availability to get the data back to and dynamic data from one to another. So we're not doing the old days of sticking with. You've got this much bandwidth, you've got this much, but it really needs to be based on an operational perspective and that dynamics got to change. So as we shift between operations on the National Mall, how do you push that where you need feeds back to a Secret Service element, cbp, tsa I'm sure FEMA's there too somewhere, but I think we're still looking for that partnership to understand how can you start doing that better and better for the dynamic bandwidth thing.

Craig Basham: 16:30

But also, how do you push the right data to the right time so you're not sitting there with a technician in the back, you know, ramping it up, ramping it down and throttling all that data as we go? So that's one area that we're looking at right now. As a matter of fact, there's RFI, so really could use your help in thinking about that, looking at the technology that we can bring in from camera systems, tactical communications, mission critical push to talk apps, you name it. We kind of are looking. This is like the perfect scenario to pull both the requirement in from the federal side and the vendor's expertise on this area. That's critical to the government, critical to the nation. And how do you get the right solution in time and securely?

Dr. Dan Gerstein: 17:18

Good and if you were characterizing kind of by way of a percentage how far along you are on this journey.

Craig Basham: 17:29

We've drawn it. No, we've gotten some proposals in. Probably 20% of the planning is done, that we're looking at. A lot of the components already have pieces there, a lot of towers there, the cameras are there, but how do you tie in that back end?

Dr. Dan Gerstein: 17:50

Great. How about turning that to FEMA and let me see what FEMA's thoughts are on? And don't be bashful by giving us a percentage as well.

Charlie Armstrong: 17:58

Yeah. So I mean I've got to back up to where we really are and that is the cloud journey right. And so there has been a fair amount of work in some of the programs on getting to cloud, certainly the pivot program moving out into GCP, gmm, modernization being built in the cloud. Currently some of the individual assistance development is being done in cloud. And then Ms Hutchinson, prior to my arrival, started an effort to really start moving us out of the existing data centers and we've got two kind of tracks going on there where we're doing some lift and shift of systems and then some replatforming, and so the lift and shift work has been going on for a little over a year now it's about 48% along and the the replatforming work we've just really gotten the point where we've identified the first five systems that we're going to replatform and hopefully start migrating some of those workloads in the next couple of months. So that's kind of where we are in cloud.

Charlie Armstrong: 19:21

I really believe we have to get through some of those big hurdles before we can really do a good design around where we want to go with edge, but certainly when you look at things like flood insurance, flood mapping, where they want to head with direct-to-customer.

Charlie Armstrong: 19:44

I believe there's probably some discussions that could start occurring now with like the Sam Holtzman's of the world on how that could fit into where they want to head. If I were to kind of read the crystal ball, I would say that's probably one of the first areas that will probably go after some edge computing and start to tackle that part of their mission, because they do a lot of work that's not just at the primary field office that gets stood up out there in a disaster, they're out in the communities and working door to door in a lot of cases. Much more need for ubiquitous computing and cannot rely on cell networks that may be down for weeks at a time while terrestrial infrastructure is being stood back up in order to reconstitute that stuff. I would say those are the first areas I would see. So if I had to give it a percentage, I would say less than 5% there, okay.

Dr. Dan Gerstein: 20:56

I'll get to you.