Metabase turns data in connected databases into charts, saved questions and shared dashboards. It suits teams that want colleagues to explore business data without writing SQL for every question, while still giving analysts a SQL editor. Self-hosting keeps operation of the analytics service with your team; the databases it queries can live elsewhere.
Requirements last checked against the project's documentation.
A sensible first server for Metabase is 2 vCPU, 4 GB RAM and 25 GB of storage.
The cheapest listed plan that fits is OVHcloud VPS-1 (US) at $4.54/month.
Dashboard complexity, simultaneous viewers and query duration affect the deployment. A slow source query can remain slow after adding memory to Metabase. Check the source database and query plan before treating a larger analytics server as the solution.
Where to start
A sensible first server for Metabase
vCPU
2
RAM
4 GB
Disk
25 GB
Metabase's production guide sets a baseline of 1 core and 1 GB of RAM, plus 1 CPU and 2 GB for every 20 concurrent users; we rounded that up to 2 vCPU and 4 GB. Disk is our estimate.
One cheapest plan per provider, based on each provider's standard list price. Check the plan page for introductory and renewal pricing before you buy.
Deploy with
Supported paths
Docker container
The host must provide
Runtime
Application database
PostgreSQL for the production setup; follow the compatibility instructions for your Metabase release.
Storage
Persistent data
Metabase’s application database stores dashboards, questions and configuration. It is separate from the business databases being analysed. Back up that application database as well as protecting the original data sources.
Step by step
Plan your Metabase deployment
Prepare the application database
Use a persistent production database and supply its connection settings to the container.