Grafana Mimir
Grafana Labs' horizontally scalable, multi-tenant Prometheus-compatible long-term metrics store.
| License | AGPL-3.0 |
| Repository | https://github.com/grafana/mimir |
| GitHub stars (approx) | 5000 |
| Language | Go |
| Self-host deployment | kubernetes helm, jsonnet, docker |
| Commercial hosted option | yes (vendor-run) |
| Freshness status | Fresh |
| Freshness checked | 2026-05-17 |
| Categories | metrics |
| Replaces SaaS | datadog, new-relic |
Mimir is the successor to Cortex (the original distributed Prometheus). It targets billion-active-series workloads with multi-tenancy, object-storage backing, and PromQL compatibility.
Pick it if: you’re at the scale where vanilla Prometheus federation breaks down (>10M active series), you live inside the Grafana ecosystem already, and you have S3/GCS/Azure Blob to back it with.
Skip it if: you’re under a few million series — Prometheus + Thanos sidecar, or VictoriaMetrics single-node, will run on a fraction of the hardware with a fraction of the operational complexity. Mimir is many microservices; respect the operational tax.
Deployment: Helm chart and jsonnet are first-class. Object storage is mandatory. Plan on a dedicated SRE for tuning at scale.