Topic

Metrics cardinality and storage cost

The operational question of how many time series a monitoring stack keeps, and what those series cost to ingest, store and query.

Current clusters

build1 publisher

Kubernetes' histogram KEP puts the migration switch in the Prometheus scrape config

KEP-5808 would expose Kubernetes component metrics as Prometheus native histograms alongside the classic bucket series, but only to scrapers that negotiate a protobuf format, and removing the old series is an explicit non-goal.

Publishers:kubernetes.dev

Reality

Evidence58
Adoption15
Hype gap+22
Incentives40
Confidence62