Kubecost · Rate Limits

Kubecost Rate Limits

Kubecost's monitoring APIs (Allocation, Assets, Cloud Cost, Budget, Forecast, Savings) are served by the Kubecost cost-model in-cluster, so request rate is bound by the consumer's own deployment rather than a vendor-imposed numeric quota. Apptio does not publish per-second/per-minute API rate limits for Enterprise Cloud SaaS.

2 Limits Throttle: 429
Rate LimitingCloud CostKubernetes

Limits

Self-hosted cost-model cluster
varies
governed by self-hosted cost-model resource limits
Enterprise Cloud SaaS tenant
varies
see vendor; not publicly documented

Policies

ETL Caching
The Allocation API is integrated with the Kubecost ETL caching layer; expensive queries are cached and CSV pipelines used for large clusters.
Backoff
Use exponential backoff with jitter on 429/503 responses; honor Retry-After when present.
Window Selection
Prefer narrower time windows and aggregated steps to reduce work in the cost-model; avoid overlapping concurrent queries against the same window.

Sources