Promhub Logo Prom hub

Cluster Capacity (Karpenter)
453,918

Created 5/6/2022
Updated 5/7/2022
Revision 1
Grafana Version >=8.2.5
Datasources
Prometheus

Description

This dashboard monitors cluster capacity and efficiency across a Karpenter-managed environment, focusing on how resources are allocated and consumed by nodes and pods. It highlights node utilization, overcommitment, and residual capacity, using metrics such as karpenter_nodes_allocatable, karpenter_nodes_total_pod_requests, and karpenter_nodes_total_overhead to reveal how close the cluster is to its limits and where headroom exists for scheduling. The panels provide a real-time view of current node metrics, pod/daemon requests and limits, and overall resource pressure to support capacity planning and overcommitment decisions.

Screenshots

Dashboard screenshot 1
Source Grafana.com

Used Metrics 9

  • karpenter_nodes_allocatable

  • karpenter_nodes_system_overhead

  • karpenter_nodes_total_daemon_limits

  • karpenter_nodes_total_daemon_requests

  • karpenter_nodes_total_overhead

  • karpenter_nodes_total_pod_limits

  • karpenter_nodes_total_pod_requests

  • karpenter_nodes_total_requests

  • karpenter_pods_state

Get Dashboard✕
Download
Copy to Clipboard

Dashboard Preview - Cluster Capacity (Karpenter)

Most Grafana dashboards are not optimized for small screens. For the best experience, view this preview on a larger screen.
grafana.shipit.dev/cluster-capacity-karpenter