NVIDIA GPU metrics dashboard
878

Created 1/17/2020
Updated 1/17/2020
Revision 1
Categories
AWSDockerHost Metrics
Grafana Version >=6.5.2
Datasources
Prometheus

This dashboard displays GPU metrics collected from NVIDIA dcgm-exporter via a metric endpoint added to Prometheus. A separate endpoint is added to Prometheus via a scrape configmap as shown in the screenshot. You will need to update the Prometheus url in the datasource section for Grafana the display metrics. You can find all the steps here

Get Dashboard
Download
Copy to Clipboard
Source Grafana.com

Used Metrics 8

  • dcgm_gpu_temp

  • dcgm_power_usage

  • dcgm_sm_clock

  • dcgm_memory_clock

  • dcgm_gpu_utilization

  • dcgm_mem_copy_utilization

  • dcgm_fb_used

  • dcgm_fb_free