This is archived documentation for v1.5.x. Go to the latest version.
Grafana Enterprise Metrics is available as a pre-compiled binary, a Docker image, as well as via common OS-specific packaging. For a list of available download options, refer to the downloads page.
Get a GEM license
A valid Grafana Enterprise Metrics license token is required to run GEM’s many added features. Without a valid license token, not all of GEM’s added features will run. However, GEM will still run with all of the functionality of an open-source Cortex binary.
If you already have a license for GEM:
- From https://grafana.com, select Login.
- From the left-hand menu, select Licenses to download the license token.
If you do not yet have a license token to run GEM, contact a Grafana Labs representative.
Choose a name for your GEM cluster
GEM licenses are issued on a per-cluster basis. Each cluster of GEM that you plan to deploy requires a unique license. When we issue a GEM license, we must have a unique cluster name with which to associate the license.
A cluster name must meet the following criteria:
- is 3 to 63 characters long
- contains lowercase letters, numbers, underscores (_), or hyphens (-)
- begins with a letter or number
- ends with a letter or number
Deploy your GEM cluster
After you have a Grafana GEM license with an associated cluster name, choose a deployment method to deploy your GEM cluster:
Related Enterprise Metrics resources
Running Prometheus-as-a-service with Grafana Enterprise Metrics
Introducing Grafana Enterprise Metrics (GEM), a simple and scalable Prometheus service that is seamless to use, simple to maintain, and supported by Grafana Labs.
How Robinhood scaled from 100M to 700M time series with Grafana Enterprise Metrics
In this GrafanaCONline session, the Robinhood team tells how GME (GameStop) led to GEM (Grafana Enterprise Metrics).
Benchmarking Grafana Enterprise Metrics for horizontally scaling Prometheus up to 500 million active series
We stress-tested GEM to show how it horizontally scaled. One takeaway: Hardware usage scales linearly up to 500 million active series.