esnodecore/esnode-core repository overview

⁠ESNODE-Core

This repository contains the source, build tooling, and documentation for the ESNODE-Core Agent.

Supported server OS targets (AI infra focused)

• Ubuntu Server (primary)
• RHEL-compatible: RHEL / Rocky Linux / AlmaLinux
• NVIDIA DGX OS
• SUSE Linux Enterprise Server (SLES)
• Debian

ESNODE-Core is a GPU-aware host metrics exporter for Linux nodes. It exposes CPU, memory, disk, network, and GPU telemetry at /metrics in Prometheus text format.

⁠Features

• Single binary with zero-config defaults (0.0.0.0:9100, 5s interval).
• Collectors: CPU, memory, disk, network, GPU (NVML-based).
• Power-aware: optional power collector reads RAPL/hwmon/BMC paths for CPU/package/node power; GPU power via NVML.
• Health endpoint at /healthz.
• JSON status endpoint at /status.

⁠Quick Start (Docker)

docker run -d \
  --name esnode-core \
  --net=host \
  --pid=host \
  --privileged \
  -v /:/host:ro,rslave \
  esnodecore/esnode-core:latest

Note: Privileged mode and host PID/network are recommended for full metric visibility (GPU, RAPL, etc).

⁠Documentation

For full documentation, please visit the GitHub Repository⁠.