finds.dev← search

// the find

grafana/mimir

★ 5,133 · Go · AGPL-3.0 · updated Jun 2026

Grafana Mimir provides horizontally scalable, highly available, multi-tenant, long-term storage for Prometheus.

Grafana Mimir is a horizontally scalable Prometheus-compatible metrics backend built for long-term storage, picking up where Cortex left off. It targets teams running multi-tenant monitoring at scale — think platform teams managing dozens of product squads on a shared cluster, or anyone who's hit Prometheus's single-node ceiling. Object storage (S3, GCS, Azure Blob) handles the long-term data so you're not paying for expensive block storage.

The monolithic mode is genuinely useful: one binary, no external dependencies, and you can migrate to microservices mode later without changing your data format. The query engine parallelizes aggressively across shards and zones, so high-cardinality queries that would time out in vanilla Prometheus complete in reasonable time. Multi-tenancy is first-class — per-tenant limits, QoS controls, and data isolation are baked into the architecture rather than bolted on. The operational tooling is unusually complete: bundled Grafana dashboards, Prometheus alerts, and runbooks for the system itself, plus `mimirtool` for managing rules and alertmanager configs.

AGPL-3.0 licensing is a real adoption blocker for any commercial SaaS product — you'll either need a commercial license from Grafana Labs or careful legal review before building anything on top of this. The microservices deployment model has a lot of moving parts (distributor, ingester, querier, query-frontend, compactor, store-gateway, ruler, alertmanager) and tuning them correctly for your workload requires reading a lot of documentation and probably getting things wrong first. The ingest storage path via Kafka adds another dependency that many teams aren't ready to operate. Configuration surface area is enormous — hundreds of flags — and the interaction between per-tenant limits and global limits has historically been a source of confusing behavior that's hard to debug without understanding the internals.

View on GitHub → Homepage ↗

// want more like this?

We dig through GitHub every week and send a few repos picked for what you actually care about — each with an honest take like this one.

Get finds in your inbox → Search again →