Real-time Nvidia GPU dashboard
3 days ago
- #NVIDIA
- #Docker
- #GPU-monitoring
- Self-contained dashboard for monitoring NVIDIA GPUs on remote servers via browser without SSH.
- Runs in a single container on one port with no configuration required.
- Provides 7 charts per GPU including utilization, temperature, memory, power draw, fan speed, clock speeds, and power efficiency.
- Features automatic multi-GPU detection, GPU process tracking, and system CPU/RAM monitoring.
- Supports threshold indicators for temperature, utilization, and memory.
- Collects core and advanced metrics like PCIe generation, performance state, compute mode, and throttle status.
- Easy setup with Docker and NVIDIA Container Toolkit.
- Optional customization available via environment variables and code modifications.
- Includes debugging tips for GPU detection issues and encourages community contributions via GitHub.