grafana/mimir

Grafana Mimir is an open source system for storing and querying large amounts of metrics data over long periods of time. It is designed as an extension to Prometheus, a widely used tool for collecting and monitoring application and infrastructure metrics. While Prometheus works well on a single machine, it has limits on how much data it can hold and how many metrics it can track at once. Mimir removes those limits by running as a distributed system across multiple machines. The core benefits described in the README include the ability to handle up to one billion active time series at once, storage on low-cost object storage services like Amazon S3 or Google Cloud Storage, and the ability to query metrics from multiple Prometheus instances in one place. It also supports running multiple isolated tenants on the same cluster, which is useful for organizations where different teams or departments each need their own separate view of their data. Mimir is built to stay available even when individual machines fail. It replicates incoming data so nothing is lost during hardware problems, and it can be upgraded or restarted without any downtime. For getting started, Mimir can run as a single binary with no external dependencies, which simplifies initial setup. For larger production deployments, it can be split into separate components that scale independently. The project ships with pre-built monitoring dashboards and alerts to help operators keep an eye on the system itself. The project is maintained by Grafana Labs and is released under the AGPL-3.0 open source license.