Grafana Mimir
Grafana Mimir is a horizontally scalable, highly available, multi-tenant time series database that provides long-term remote-write storage for Prometheus and OpenTelemetry metrics. It began as a fork of Cortex and is designed to scale metrics storage well beyond what a single Prometheus instance can retain.
Baseline topologies
| Topology | Minimum nodes | Roles |
|---|---|---|
| monolithic | 1 | mimir |
| microservices | 6 | mimir-distributor, mimir-ingester, mimir-compactor, mimir-store-gateway, mimir-querier, mimir-query-frontend |
What drives the cost
- Active time series
- Total unique active time series being ingested. Primary driver for ingester memory.
- Ingestion rate (samples/second)
- Sustained metric sample ingestion rate. Drives distributor and ingester CPU.
- Retention period (months)
- How long to retain metric data. Drives object storage capacity via store-gateway.
Where it runs
bare_metal, dedicated, cloud_vm
Size Grafana Mimir and get a cost estimate