Getting Started
Lakemeter is a Databricks App — a managed web application with built-in SSO authentication that runs entirely on the Databricks platform.
Prerequisites
You need a Databricks CLI configured with a workspace profile. All other permissions (Lakebase, secret scopes, Apps, serverless compute) are granted to workspace users by default.
If you can't install the Databricks CLI locally, you can run the installer directly from your workspace. Create any notebook on a serverless cluster, click the terminal button (bottom-right corner), and use the pre-installed CLI — no profile needed since it's already authenticated.
Click the terminal button in the bottom-right corner of any notebook.
The Databricks CLI is pre-installed and authenticated in the notebook terminal.
# In the notebook terminal — CLI is pre-installed and authenticated
git clone <repository-url>
cd lakemeter-oss
./scripts/install.sh --non-interactive
Install
git clone <repository-url>
cd lakemeter-oss
./scripts/install.sh --profile <your-cli-profile>
The installer provisions everything automatically in ~15 minutes: Lakebase instance, database schema, pricing data, app configuration, and deployment. See the Installer Guide for the full walkthrough.
For a list of all resources created by the installer, see the Deployment Inventory.
After Installation
Once the installer completes, your app is live at:
https://lakemeter-<workspace-id>.<cloud>.databricksapps.com
Users access the app through their Databricks workspace — authentication is handled automatically via SSO. No additional user setup is required.
Updating
To update Lakemeter after a new release:
git pull
./scripts/install.sh --profile <your-cli-profile>
The installer is idempotent — re-running it updates pricing data and redeploys the app without losing existing estimates.