The push to modernize government IT is colliding with a hard reality: The most important data isn’t always generated where networks are strong, cloud connectivity is reliable, or power and cooling are abundant. Increasingly, missions demand compute that can run wherever the work happens – at remote facilities, tactical environments, and even in orbit –while still meeting security, cost, and resilience requirements that don’t bend just because the edge is hard.
That was the message from a Red Hat Government Symposium session, “Edge to Enterprise: Scaling Sustainability Across Civilian and DOD Missions,” moderated by Travis Steele, chief architect for Air Force and Space Force at Red Hat. Steele was joined by Baibhav “Bobo” Devkota, executive director, Border Enforcement and Management Systems Directorate at U.S. Customs and Border Protection (CBP), and Dennis Gatens, president of LEOCloud at Voyager Technologies, a national security and space solutions company.
Devkota and Gatens pointed to the same strategic shift: Agencies are treating edge computing less like a niche deployment model and more like a way to scale mission outcomes without scaling cost and risk.
Cutting bandwidth and cloud spend by filtering at the source
For Devkota, the edge is increasingly where CBP can make the economics of modern sensing work – particularly with video.
He described the sheer operational scale that CBP is grappling with: “tens, hundreds of video feeds, 4K, 1080p, a mismatch of all of them.”
That type of data volume forces architectural decisions that quickly become budget decisions. Shipping every frame to a centralized cloud environment isn’t feasible – it takes too long and costs too much. CBP processes video feeds closer to where they are collected, so it can “just send what’s interesting in those video feeds back up,” Devkota said.
For CBP, edge computing is driven by a convergence of three primary objectives: controlling transport and cloud costs, reducing latency for operators who need timely cues, and enabling analytics that can run in constrained environments. “It’s a financially good play,” Devkota said, and a “mission optimization play to do all three.”
Moving the edge to space
While many government edge conversations start with remote sites or tactical nodes, Gatens offered a different starting point: “The edge, from our perspective, is in space.”
Space-based compute is an extreme version of constraints that are becoming familiar on Earth – latency, limited transport, intermittent connectivity, and the need to maintain trust in data while pushing processing closer to collection.
“One of the things we look at, as we see the natural evolution of cloud extending into space, is the ability to process data, because it is about security, latency, and transport,” Gatins observed. “As communications infrastructure evolves in space into these layers of mesh network capabilities, you can reach anywhere, at any time, with a high level of urgency, with insight to an end user that has an urgent need for guidance.”
In September 2025, Voyager Technologies’ LEOcloud Space Edge was deployed to the International Space Station to demonstrate cloud computing infrastructure on the space station and its ability to integrate with terrestrial networks.
“We are a managed service in space today … offering it as a sandbox to begin realizing the benefits of having your applications and your insight derived in space and delivered directly to the end user anywhere on Earth,” Gatens said.
Making resilience an architectural requirement
Only a few agencies are headed to orbit, but the operating principles for edge computing in space also apply to edge computing on Earth, Steele observed: Build for disruption by default, assume transport is limited, and ensure that systems can grow, update, and remain secure over time as mission demands change.
At the edge, resilience becomes a core design objective. Remote nodes still need to be patched, managed, and defended. They need to operate even when connectivity is intermittent, bandwidth is constrained, or the environment is physically exposed.
For CPB, edge computing means dispersed deployments across ports of entry, remote checkpoints, and other facilities. For defense and space missions, it can mean environments where downtime and delayed updates create immediate mission risk.
“The ability to continuously operate, no matter the environment, is critical,” Steele said.
In other words, Gatens observed, “You can’t roll a truck. So you better be able to allow that system to autonomously take care of itself, heal itself, or … have agility built in to reach it and manage it.”
Designing for security
As edge deployments expand, so does the attack surface, especially when systems are deployed outside controlled data centers. At CBP, for example, “Some of our facilities are physically under control. Some of them are sitting wide open on the border,” Devkota said. “So we do [a] security assessment every time we design or implement … at the edge.”
That physical reality shapes how agencies design identity, access, device trust, and monitoring. It also influences the workforce, because edge environments require people who can translate high-level security strategy into secure system design.
“In our world,” Devkota observed, “we need almost as many security architects as ISSOs (information system security officers) to make sure that we design things securely and then implement them likewise.”
Moving from pilots to repeatable architectures
Edge deployments often begin as pilots tied to a specific mission need – a sensor rollout, a field operations requirement, a surveillance deployment, or a connectivity workaround. The challenge for government IT leaders is scaling those wins without creating an unmanageable patchwork of bespoke systems.
The takeaway from Steele, Devkota, and Gatens was that “edge to enterprise” is a maturity curve. Agencies are moving from isolated edge experiments to repeatable architectures that can be governed like enterprise platforms while meeting the needs of field operators and mission owners. It’s less about adopting edge computing as a category and more about adopting the practices that make it sustainable: process data where it’s created, transport only what matters, build for disruption, engineer security into the design, and operate remotely as a default.
Watch the Red Hat Government Symposium session: “Edge to Enterprise: Scaling Sustainability Across Civilian and DOD Missions,” and explore more sessions from the Red Hat Government Symposium.