NVIDIA Fleet Intelligence Host Agent

This Helm chart deploys the NVIDIA Fleet Intelligence Host Agent as a Kubernetes DaemonSet and provides the OpenTelemetry gateway used to forward telemetry to a configured backend. It is intended for operators monitoring GPU health and system telemetry across Kubernetes GPU nodes.

Features

• Deploy the host agent on selected nodes using a DaemonSet.
• Configure GPU and host telemetry collection and the telemetry gateway.
• Configure optional Fleet Intelligence enrollment using Kubernetes Secrets.
• Customize node selection, tolerations, resource settings, and image references through Helm values.

Requirements

A Kubernetes cluster with Helm, supported Linux GPU nodes, NVIDIA drivers, and a reachable standalone DCGM HostEngine is required for GPU collection. Follow the installation guide for the NVIDIA GPU Operator and DCGM configuration. The cluster must be able to pull the agent and gateway images. Preserve agent state at /var/lib/fleetint and provide backend connectivity and enrollment credentials when remote enrollment is enabled.

Installation and configuration

Follow the Helm installation guide for prerequisites, enrollment, installation, and upgrades. Consult the chart configuration reference for available values.

Publisher and source

Published by NVIDIA. Source code and contribution history are available in the Fleet Intelligence Agent repository.

License

The Fleet Intelligence Agent project source is licensed under the Apache License 2.0. Container dependencies retain their respective licenses.