Datadog Dashboards

NVIDIA DCGM Dashboard for Kubernetes (MIG & Non-MIG GPUs)

Good 930 downloads0 views0 commentsRevision 1Created Converted from Grafana dashboard #23382 by M.Mahdi Setak
How to import

In Datadog: Dashboards → New Dashboard → ⚙ Configure → Import dashboard JSON → paste this JSON.

Reactions

This Grafana dashboard uses the NVIDIA DCGM exporter to monitor GPU performance in Kubernetes clusters. Supporting both MIG and non-MIG GPUs, it provides real-time metrics on utilization, memory, temperature, power, and more, enabling efficient management of NVIDIA GPU resources.

NVIDIA DCGM Dashboard for Kubernetes (MIG & Non-MIG GPUs)

Conversion quality

Good 93

Share of panels whose queries were translated without loss. Partial and unsupported panels keep the original PromQL in a note widget.

  • Native0
  • OpenMetrics17
  • Partial3
  • Unsupported0

Metrics without a native Datadog mapping

These metric names were kept in OpenMetrics naming. Adjust them if you collect with a native Datadog integration.

DCGM_FI_DEV_FB_USED ×4DCGM_FI_DEV_SM_CLOCK ×3DCGM_FI_PROF_GR_ENGINE_ACTIVE ×2DCGM_FI_PROF_PIPE_TENSOR_ACTIVE ×2DCGM_FI_PROF_DRAM_ACTIVE ×2DCGM_FI_DEV_TOTAL_ENERGY_CONSUMPTION ×2DCGM_FI_DEV_FB_FREEDCGM_FI_DEV_GPU_TEMPDCGM_FI_DEV_POWER_USAGEDCGM_FI_DEV_MEM_CLOCKDCGM_FI_DEV_MEMORY_TEMPDCGM_FI_DEV_NVLINK_BANDWIDTH_TOTALDCGM_FI_DEV_PCIE_REPLAY_COUNTERDCGM_FI_DEV_XID_ERRORS
Tagsgpukubernetes-mixinmig

Revisions

  • Revision 1Converted from grafana.com revision 1Download JSON

Rate this dashboard

Sign in to rate.

0 comments

No comments yet.