Skip to content
Documentation for the on-premise version of PM2 Enterprise
Branch: master
Clone or download
Type Name Latest commit message Commit time
Failed to load latest commit information.
docker docker: add our nginx conf Mar 14, 2019
terraform-gce-k8s fix: remove abac forced, create serviceaccount for tiller Mar 11, 2019
.drone.yml docker: add our nginx conf Mar 14, 2019
LICENSE doc: improv baremetal documentation Mar 7, 2019

PM2 Enterprise On-Premise

PM2 Enterprise logo


We, at Keymetrics, have heavily invested for the past year in building a product that can be scaled easily and launch on-premise in any infrastructure.

To use our product on-premise, we provide few docker images containing all of our applications, from backend services to frontend webapps.

You only need to create a account on, we allow this account to pull our images so you can install them in your servers.

If you have any questions :


To make it really easy to deploy our product, we also give you examples to deploy on different infrastructures:

Don't hesitate to ask questions to our tech team, we can provide assistance if you want to deploy it on another providers (like AWS).

Ressources requirements

And in terms of the requirements itself, it really depend of how much agents you will need to connect but if we aim for 500 agents in average (maximum ~1000 in burst):

  • Elasticsearch (that would be the biggest VM that you will need, we advise to use 16G of RAM for it and at least 4 CPU and a good persistent storage, if possible SSD)
  • Redis (at least 4G of RAM and 1 CPU)
  • Mongodb (500Mb of RAM and 1 CPU is enough and persistent storage)
  • For the applications themselves, we advise at least 10G of RAM and 12 CPU for all applications.

Note that it really depend on your use of the tool, we recommend to ask the technical team for a better estimate that suit your needs. Also, this is an example, for the same load you could use the same server for both applications and the redis server.

How it looks with a graphic

Here are the most basic deployment with docker-compose should looks like:


Note that in this example we put the redis instance on the same server as the applications.


You can’t perform that action at this time.
You signed in with another tab or window. Reload to refresh your session. You signed out in another tab or window. Reload to refresh your session.