OpenShift Project Quota Calculator

requests = projects × pods × request per pod (the figure the scheduler reserves); limits = the same with limits; the overcommit ratio is limit ÷ request. ResourceQuota governs the sum of requests and limits a project may hold, and the scheduler only ever counts requests. That makes the limit-to-request ratio the real risk dial: a ratio of 1 gives Guaranteed QoS with no bursting, while a high ratio packs nodes densely and relies on workloads never bursting together.

Quotas set from limits rather than requests either strand capacity or oversubscribe nodes into OOM kills, and quotas totalling more than allocatable capacity produce unschedulable pods with no obvious cause.

This calculator takes 8 inputs: Projects with quota, Pods per project, CPU request per pod, CPU limit per pod, Memory request per pod, Memory limit per pod, Allocatable cluster vCPUs, Allocatable cluster memory. The pre-filled defaults are a realistic starting point — replace them with figures from your own environment for a result you can act on.

A ResourceQuota caps the project total but rejects pods that specify no request at all. A LimitRange supplies the per-pod defaults, so in practice you need both — the LimitRange to make unspecified pods admissible and the quota to bound the sum.