Skip to main content
Version: 6.2

AI Observability: GPU Status

The AI Observability: GPU Status dashboard shows the status of GPUs serving inference workloads. It tracks GPU load, memory usage, temperature, and hardware errors to identify signs of equipment degradation.

Useful for: ML infrastructure engineers, system administrators, DevOps.

Data source: gen_ai_gpu_metrics*

Main Sections​

1. AI Summary and General GPU Parameters​

Shows a summary of GPU status and information about each GPU:

  • AI summary: overall GPU status, a brief assessment, identified issues, and recommendations
  • model, UUID, PCI address, server, and driver version
  • total and used GPU memory

General Parameters and GPU Status

2. GPU Status​

Shows the main GPU operating metrics:

  • current GPU load
  • core temperature and power consumption
  • presence of XID errors
  • load, temperature, power, and error trends over time

3. GPU Memory​

Shows the status and dynamics of GPU memory usage:

  • GPU memory activity
  • GPU memory temperature
  • GPU memory frequency
  • GPU memory utilization
  • memory metric trends over time

GPU Memory