Fleet Command is a hybrid-cloud software platform for deploying and managing AI systems and applications at the edge. Its features enable you to:

• Manage edge systems in multiple locations with a single control plane—You can pair PowerEdge servers in multiple locations with Fleet Command, which allows you to deploy a complete operating environment and application software stack on those servers. Groups of servers in a specific edge location are managed as a single Kubernetes cluster under the control of Fleet Command.
• Deploy applications from private or public catalogs—Fleet Command allows you to deploy applications from the public NGC catalog and from your NGC Private Registry to the edge systems. Organizations can use applications already available in the public catalog, develop and deploy in-house AI applications, and store them in their private registry. Administrators can then push these applications to edge locations as Helm charts so that organizations can take advantage of consistent, secure, and reliable environments to accelerate development-to-production cycles.
• Connect safely to systems with remote management—You can track system status to ensure that systems are ready to run applications. Systems that are accessed remotely are marked as having been accessed using a remote console. If the system is marked as accessed but you do not know who accessed it, you can reimage the system to alleviate security concerns.

Fleet Command enables remote console access to edge systems for troubleshooting. It also provides centralized access to all the system logs, including Kubernetes and Linux system logs.

• Secure deployments—Fleet Command has integrated, end-to-end security to ensure that AI applications and data are always protected:

• Applications are scanned for vulnerabilities and malware before they are loaded. Signed containers ensure that only authenticated software is deployed to the edge.
• Storage in the local drives is encrypted.
• Secure Boot and Measured Boot provide a trusted boot process.

• Complete the life cycle (for edge systems and applications)—Fleet Command provides an end-to-end life cycle for both the edge systems and the applications. New edge systems can be deployed and then kept current through over-the-air updates or unneeded systems removed with the Fleet Command stack. Also, applications can also be deployed and updated from the same Fleet Command console.

Fleet Command also handles all the once-complex "day two" management tasks like updating system software over the air or monitoring location health at all your edge locations from a central control plane.

• Provide resiliency—Fleet Command takes advantage of Kubernetes capabilities to provide a resilient software stack that allows all systems to self-heal when applications are disrupted. Multiple nodes in a single location form a Kubernetes cluster.

For a comprehensive list of capabilities, see the NVIDIA Fleet Command documentation.

The Fleet Command software stack includes Ubuntu, Kubernetes, Helm, Tiller, and tools for deploying and managing the NVIDIA hardware assets necessary for GPU-enabled Kubernetes. With a few clicks, the software stack is automatically installed on any targeted edge servers. Fleet Command configures and manages the remote systems through an abstraction layer that shields customers from the complexity of full edge software stack management.