Home / BeaverDeck / Docs / Insights Guide / Node Insights / Node Underutilization

Node Underutilization

BeaverDeck uses this check to identify a specific node condition that may need operator review.

Permissions: viewing checks requires insights: view. Opening a linked object or logs requires the corresponding resource permission, and the BeaverDeck ServiceAccount must be allowed to read the Kubernetes resources used by the check. Suppressing a finding requires insights: edit and affects all users.

Check type	`node-underutilized`
Insights section	Node Insights
Alert severity	Warning

When It Reports A Finding

On a schedulable node with known allocatable resources, combined scheduled pod CPU requests and memory requests are both below 50% of allocatable capacity.

Why This Is A Problem

Persistently low requested utilization can indicate avoidable node cost, uneven placement, oversized node pools, or requests that do not reflect actual workload demand.

Recommended Response

Compare the request-based signal with longer-term CPU and memory usage before changing capacity.
Review pod requests, affinity, topology constraints, taints, and disruption budgets that may prevent consolidation.
Right-size requests or node pools, and use cluster autoscaling or a controlled consolidation process where appropriate.

Scope And Limitations

This is based on requests, not real-time utilization. Intentional failover headroom, reserved system capacity, topology requirements, and burst capacity can make the warning acceptable.

After remediation: refresh Node Insights and verify the underlying resource or metric. Suppress the finding only when the condition is intentional and its risk is accepted.