README
¶
NVIDIA Fleet Intelligence Agent
NVIDIA Fleet Intelligence Agent - Host agent for GPU telemetry collection and attestation.
Built on top of leptonai/gpud
Overview
What It Monitors:
- GPU Metrics: Power, temperature, clocks, utilization, memory, Xid events
- System Metrics: CPU, memory, disk, network usage
- Infrastructure: NVIDIA drivers, CUDA runtime, InfiniBand, containers
Export Formats:
- HTTP API Server: Serves data via REST endpoints (JSON) and Prometheus metrics (
/metrics) - File Export (Offline Mode): Writes data to local files in CSV or JSON format
- Remote Export: Sends telemetry data to OpenTelemetry-compatible endpoints via OTLP over HTTP
Key Features:
- Lightweight: <500MB RAM, <1% CPU usage
- Non-intrusive: Read-only operations, no system modifications
- Production-ready: 24/7 datacenter operation
Supported Platforms
| OS Family | Supported Versions | Architecture | GPU |
|---|---|---|---|
| Ubuntu | 22.04, 24.04 | x86_64, ARM64 | Ampere, Ada Lovelace, Hopper, Blackwell, Rubin |
| RHEL | 8, 9, 10 | x86_64, ARM64 | Ampere, Ada Lovelace, Hopper, Blackwell, Rubin |
| Rocky Linux | 8, 9, 10 | x86_64, ARM64 | Ampere, Ada Lovelace, Hopper, Blackwell, Rubin |
| AlmaLinux | 8, 9, 10 | x86_64, ARM64 | Ampere, Ada Lovelace, Hopper, Blackwell, Rubin |
| Amazon Linux | 2023 | x86_64, ARM64 | Ampere, Ada Lovelace, Hopper, Blackwell, Rubin |
Documentation
Important: Documentation links are relative to the branch or tag you are viewing. The default GitHub view uses main, which may describe unreleased changes. When installing or upgrading a specific agent version, switch to that version's release tag first.
- Helm Installation - Kubernetes (Helm) installation and troubleshooting
- DEB Installation - Ubuntu package install, update, and uninstall
- RPM Installation - RHEL/Rocky/Alma/Amazon package install, update, and uninstall
- Architecture - Bare metal and Kubernetes architecture, dependencies, and runtime flow
- Usage - Commands, HTTP API, integration, and troubleshooting
- Configuration - Environment variables and service configuration
- Development - Building from source and contributing
Contributing
See CONTRIBUTING.md for development setup and guidelines.
Related: leptonai/gpud (upstream dependency)
License
Apache License 2.0 - see LICENSE for details.
Directories
¶
| Path | Synopsis |
|---|---|
|
cmd
|
|
|
fleetint
command
|
|
|
internal
|
|
|
agentstate
Package agentstate centralizes access to local persisted agent state.
|
Package agentstate centralizes access to local persisted agent state. |
|
attestation
Package attestation owns the backend attestation workflow.
|
Package attestation owns the backend attestation workflow. |
|
backendclient
Package backendclient provides the agent-facing client for backend workflows.
|
Package backendclient provides the agent-facing client for backend workflows. |
|
enrollment
Package enrollment provides shared enrollment functionality for the Fleet Intelligence agent.
|
Package enrollment provides shared enrollment functionality for the Fleet Intelligence agent. |
|
exporter
Package healthexporter provides functionality to export health data from local SQLite to a global health endpoint for centralized monitoring and long-term storage using OTLP format.
|
Package healthexporter provides functionality to export health data from local SQLite to a global health endpoint for centralized monitoring and long-term storage using OTLP format. |
|
exporter/collector
Package collector handles health data collection from various sources
|
Package collector handles health data collection from various sources |
|
exporter/converter
Package converter handles conversion of health data to different formats
|
Package converter handles conversion of health data to different formats |
|
exporter/writer
Package writer handles writing health data to various outputs
|
Package writer handles writing health data to various outputs |
|
inventory
Package inventory owns inventory collection and sync state.
|
Package inventory owns inventory collection and sync state. |
|
inventory/mapper
Package mapper contains inventory payload mappers.
|
Package mapper contains inventory payload mappers. |
|
inventory/sink
Package sink contains inventory sink implementations.
|
Package sink contains inventory sink implementations. |
|
inventory/source
Package source contains inventory collection adapters.
|
Package source contains inventory collection adapters. |
|
machineinfo
Package machineinfo provides a shim layer over gpud's machine-info package to customize version information for Fleet Intelligence.
|
Package machineinfo provides a shim layer over gpud's machine-info package to customize version information for Fleet Intelligence. |
|
nodeidentity
Package nodeidentity initializes and persists the stable node UUID used by enrollment, inventory, attestation, and metrics workflows.
|
Package nodeidentity initializes and persists the stable node UUID used by enrollment, inventory, attestation, and metrics workflows. |
|
precheck
Package precheck evaluates prerequisite checks for enrollment and installation flows.
|
Package precheck evaluates prerequisite checks for enrollment and installation flows. |
|
registry
Package registry provides component registration and management for fleetint, allowing fine-grained control over which components are enabled.
|
Package registry provides component registration and management for fleetint, allowing fine-grained control over which components are enabled. |
|
scan
Package scan provides system scanning functionality for Fleet Intelligence monitoring.
|
Package scan provides system scanning functionality for Fleet Intelligence monitoring. |
|
server
Package healthserver provides a simplified HTTP server for Fleet Intelligence metrics export.
|
Package healthserver provides a simplified HTTP server for Fleet Intelligence metrics export. |
Click to show internal directories.
Click to hide internal directories.