grafana/mimir

Project Overview

Grafana Mimir is an open-source, horizontally scalable time-series database built for long-term storage of Prometheus metrics. It lets teams aggregate metrics from many Prometheus instances and query them globally, using object storage such as S3, GCS, Azure Blob Storage, OpenStack Swift or any S3-compatible store for durable retention.

It is natively multi-tenant, with tenant isolation and limits, and replicates incoming metrics for high availability. The project claims internal testing at up to 1 billion active time series, and its architecture supports restarts and upgrades without interrupting ingestion or querying. It suits platform and observability teams replacing a single Prometheus instance or migrating from Thanos, Prometheus or Cortex.

Getting Started

For a first deployment, start with the official Get started guide. For production, the documentation recommends reading the architecture overview, configuration guide, and production runbook before deploying:

• Architecture overview
• Configuration
• Run Mimir in production

Deployment instructions are in Deploy Grafana Mimir. The README states that monolithic mode runs with a single binary and no additional dependencies, but it does not provide the binary download URL or CLI flags directly; use the cited deployment documentation for exact commands.

For migrations, use the specific guides for Thanos or Prometheus and Cortex.

Usage Constraints

The README does not list the minimum Go version, supported operating systems or resource sizing. Those details belong in the deployment and production documentation. Long-term storage requires an object store, and the available backends are AWS S3, Google Cloud Storage, Azure Blob Storage, OpenStack Swift and S3-compatible storage.

Mimir supports multi-tenancy with per-tenant limits and quality-of-service controls. Production readiness requires reading the architecture and configuration material; the repository is distributed under AGPL-3.0-only.

HotRepo's Take

The only recorded HotRepo snapshot is a daily Go-language list entry on 2026-11-06 at position 12, with 2 stars gained that day; the project had 5,277 stars and 867 forks at the time. There is no all-language daily rank and no captured star curve, so this single record is not enough to describe a trend. The project's relevance rests on its documented role as scalable, multi-tenant Prometheus storage rather than on that snapshot.