BeaverDeck logo
BeaverDeck

Home / BeaverDeck / Docs / Insights Guide / GPU Insights / GPU Quota

GPU Quota

BeaverDeck uses this check to identify a specific gpu 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 typegpu-quota
Insights sectionGPU Insights
Alert severityWarning

When It Reports A Finding

A selected namespace has active GPU requests but no ResourceQuota hard limit for requests.nvidia.com/gpu, limits.nvidia.com/gpu, or nvidia.com/gpu.

Why This Is A Problem

Without a namespace quota, one tenant or workload group can consume all available GPUs and make capacity allocation less predictable.

Recommended Response

  1. Determine the namespace's approved GPU allocation and burst policy.
  2. Create or update a ResourceQuota using the GPU resource key supported by the cluster.
  3. Coordinate quota values with node capacity, workload requests, and organizational ownership.

Scope And Limitations

A quota is a policy choice, not universally required. Dedicated single-tenant namespaces or clusters may intentionally omit it.

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