GCP – Google Distributed Cloud at the edge powers U.S. Air Force Mobility Guardian 2025

For today’s mission owner, operating effectively in denied, degraded, intermittent, and limited bandwidth (DDIL) environments is paramount. The Department of Defense’s strategy requires smaller, dispersed teams to function autonomously, creating a critical need for secure AI and data processing at the edge.

Bringing cloud power to the edge

Recognizing this challenge, the U.S. Air Force partnered with Google Public Sector and General Dynamics Information Technology (GDIT) to deploy the Google Distributed Cloud (GDC) air-gapped appliance. The integrated hardware and software solution offers a ruggedized, transportable appliance that allows users to securely run workloads classified up to the Secret level. GDC’s secure-by-design architecture ensures that Zero Trust security is integral, not an afterthought, minimizing complex configurations in contested environments.

GDC at Mobility Guardian 2025

This July in Guam, The United States and allied partners participated in Mobility Guardian 2025, Air Mobility Command’s (AMC) premier readiness exercise part of the Air Force’s Department-Level Exercise (DLE) series. During this key test of modern expeditionary capabilities, the ruggedized GDC air-gapped appliance successfully demonstrated its ability to generate critical intelligence and AI-powered insights in a disconnected environment – which is fundamental to enhancing resilience and accelerating decision-making for the Air Force.

Delivering a decisive advantage: key capabilities in action

During Mobility Guardian, the Google Distributed Cloud appliance successfully delivered:

• Resilient command and control: Provided a secure, disconnected IL2 collaboration platform with a containerized version of MatterMost, an open-source platform designed for secure communication. Airmen used Google’s generative AI models for real-time transcription, optical character recognition (OCR), translation, and summarization to improve interoperability.
• Real-time data processing: The appliance integrated with GDIT’s Luna AI solution to process unified air defense data at the edge. This provides a low-latency picture for real-time flight path tracking and improved situational awareness which can be leveraged for counter-UAS triggers.
• Edge development environment: The team provisioned a coding environment with Jupyter notebooks that engineers could use to develop and refine applications directly in the field, enabling the creation of new analytics and tools at the tactical edge.
• AI-enabled tele-maintenance: The Air Force ingested hundreds of pages of technical manuals to create a grounded generative AI resource for instant troubleshooting. Its computer vision capabilities analyzed visual data from the field to automatically identify and extract critical elements – such as maintenance abnormalities, equipment status, or location coordinates – turning unstructured information into actionable readiness data.

From the edge to the enterprise: validating the future of defense operations

This collaboration between the Air Force, Google, and GDIT exemplifies how tactical, air-gapped cloud systems can bring out-of-the-box GPUs and AI-enhanced solutions to the mission – whenever and wherever they’re needed. This successful demonstration also validated a complete workflow – from operating autonomously at the edge to syncing seamlessly back to the core network once connectivity was restored. The insights gained are invaluable for the future of defense operations. The GDC air-gapped appliance is now ready for deeper integrations with mission applications and connectivity with Google’s Global Network (GGN) to create a secure high performance link from the edge back to the cloud.

To learn more about how we are empowering defense and intelligence agencies, register to attend the Google Public Sector Summit on October 29, in Washington, D.C.

^{Photo by Senior Airman Daniel Hernandez, 1st Combat Camera Squadron}