Grafana Mimir

active

by Grafana Labs · since 2022 · also known as Mimir

Grafana Labs' horizontally scalable, multi-tenant long-term store for Prometheus and OTLP metrics.

time-seriesdatabase engineopen source (copyleft)mature

Overview

Grafana Mimir is a horizontally scalable Prometheus backend that keeps metric blocks in object storage and adds multi-tenancy, high availability and global query federation. Forked from Cortex in 2022 and licensed AGPL-3.0, it is the engine behind Grafana Cloud Metrics. Mimir 3.0 (October 2025) introduced an ingest-storage architecture that places a Kafka-compatible log between the write and read paths so they scale independently, and made the streaming Mimir Query Engine the default, which sharply cuts querier memory. It is powerful but it is a microservices system: expect many components, a Kafka-compatible broker in the new mode, and per-tenant limit tuning.

Strengths & trade-offs

Strengths

  • +Scales to very large active-series counts across many tenants
  • +Object storage for blocks keeps long retention cheap
  • +Ingest storage separates read and write paths for predictable performance
  • +Streaming PromQL engine uses far less memory than the Prometheus engine
  • +Same code that runs Grafana Cloud Metrics

Trade-offs

  • −Many microservices to deploy, monitor and upgrade
  • −AGPL-3.0 can be a blocker for embedding or reselling
  • −New ingest-storage mode adds a Kafka-compatible dependency
  • −Overkill for a single cluster that fits one Prometheus

Great for

  • +Centralised multi-tenant metrics for many clusters
  • +Long-term Prometheus storage on object storage
  • +Platform teams offering metrics-as-a-service internally

Think twice for

  • −Small single-cluster setups
  • −Logs and traces (use Loki and Tempo)
  • −SQL analytics

Capabilities

Native or official first-party support · 4 of 25 tracked

✓TTL / expiry
✓Durable persistence
✓Compression
✓Streaming ingest
–Transactions
–Joins
–Secondary indexes
–Full-text search
–Vector search
–Geospatial
–JSON
–Change data capture
–Triggers
–Stored procedures
–Materialized views
–Pub/Sub
–Columnar storage
–In-memory
–Encryption at rest
–Role-based access
–Time travel
–Branching
–Online schema changes
–Offline sync
–Runs in WebAssembly

Scale & consistency

Comfortable data size
Massive · Multi-PB
Horizontal scaling
Native
Multi-region
No
Transactions
None
Consistency
Eventual
Schema
Schemaless
Architecture
Distributed, Shared storage
Query languages
PromQL

Distributors, ingesters, store-gateways and compactors scale independently over S3/GCS/Azure Blob.

Deployment

Self-hosted Vendor-managed cloud Serverless Embedded Edge Kubernetes operator

available on · AWS · Google Cloud · Azure

Ecosystem

compatible withPrometheus, S3
apis & protocolsPrometheus remote write, Prometheus HTTP query API, OTLP

Alternatives to Grafana Mimir

Compare all→
ThanosTime-Series13.1

Sidecar-plus-object-storage layer giving Prometheus global queries, HA dedup and long retention.

grafana-mimir vs thanos →
VictoriaMetricsTime-Series26.8

Resource-frugal, Prometheus-compatible metrics database, run as a single binary or a sharded cluster.

grafana-mimir vs victoriametrics →
M3DBTime-Series4.2

Uber-born distributed TSDB and aggregator for Prometheus-scale metrics with tunable replication.

grafana-mimir vs m3db →
PrometheusTime-Series70.8

CNCF-graduated pull-based monitoring system with a local TSDB, PromQL and alerting rules.

grafana-mimir vs prometheus →
CortexTime-Series7.2

CNCF multi-tenant, horizontally scalable long-term storage for Prometheus metrics on object storage.

grafana-mimir vs cortex →
GreptimeDBTime-Series15.1

Rust observability database that keeps metrics, logs and traces in one columnar engine on object storage.

grafana-mimir vs greptimedb →

Grafana Mimir FAQ

What is Grafana Mimir?

Grafana Mimir is a time-series database by Grafana Labs, first released in 2022. Grafana Mimir is a horizontally scalable Prometheus backend that keeps metric blocks in object storage and adds multi-tenancy, high availability and global query federation. Forked from Cortex in 2022 and licensed AGPL-3.0, it is the engine behind Grafana Cloud Metrics. Mimir 3.0 (October 2025) introduced an ingest-storage architecture that places a Kafka-compatible log between the write and read paths so they scale independently, and made the streaming Mimir Query Engine the default, which sharply cuts querier memory. It is powerful but it is a microservices system: expect many components, a Kafka-compatible broker in the new mode, and per-tenant limit tuning.

Is Grafana Mimir open source?

Yes. Grafana Mimir is open source. License: GNU AGPL v3.0.

Does Grafana Mimir support ACID transactions?

No. Grafana Mimir does not provide ACID transactions; its consistency model is eventual.

Does Grafana Mimir scale horizontally?

Grafana Mimir scales horizontally out of the box, with built-in sharding or partitioning. It is comfortable at massive scale (Multi-PB).

How can I run Grafana Mimir?

You can self-host it and use the vendor's managed cloud. Managed offerings include Grafana Cloud Metrics. It is available on AWS, Google Cloud and Azure.

What is Grafana Mimir best for?

Grafana Mimir is a strong fit for centralised multi-tenant metrics for many clusters, long-term Prometheus storage on object storage and platform teams offering metrics-as-a-service internally. Think twice for small single-cluster setups.

What are the best alternatives to Grafana Mimir?

Common alternatives to Grafana Mimir include Thanos, VictoriaMetrics, M3DB, Prometheus and Cortex. The right choice depends on your workload, scale and hosting; WhichDB's smart match ranks them against your requirements.

How popular is Grafana Mimir?

Grafana Mimir ranks #203 of 844 databases on the WhichDB Score (15.8/100), and #11 among time-series databases. The score combines public adoption signals such as package downloads, GitHub activity, job posts and developer discussion.