Grafana Mimir: High availability and scalability for Prometheus

Prometheus and Grafana are the de facto standards for metric monitoring of modern cloud native applications and infrastructures.

For organizations needing massive scale, Grafana Labs has launched Grafana Mimir as a solution for extending Prometheus, with high availability, horizontal scalability, multi-tenancy, durable storage, and blazing fast query performance over long periods of time.

Grafana Mimir is suitable for running at any scale. We’ve successfully run it on a laptop without WiFi, as well as in production with more than 1 billion active time series.

We’re excited to share this project with the open source community.

In this introductory webinar, we’ll go over:

  1. What is Grafana Mimir
  2. The Mimir architecture, requirements and configuration
  3. A live installation of Mimir
  4. A look at the cutting-edge features employed by Grafana Mimir, like query sharding and the two-stage compactor
  5. Tips to run Mimir in production, based on the experience we gained running it for some of the largest and most demanding customers
  6. The upcoming features and what you could expect from Mimir in the near future

Additional resources to explore:

Marco Pracucci

Principal Engineer, Grafana Labs

Marco is a Grafana Mimir maintainer. He contributed to building some of the Grafana Mimir core features, like blocks storage, query sharding and scalable compactor. His current focus is on further simplifying the Mimir operability at any scale.