Skip to main content

Enable Lakehouse

The Lakehouse enables you to instantly run SQL queries, build dashboards, and power AI applications directly on Atlan metadata without creating pipelines or data extracts.

This guide walks you through how to enable the Lakehouse for your Atlan workspace so you can start running queries on your Atlan metadata.

Prerequisites​

Before you begin, make sure:

  • You have the Admin or Workflow Admin role in your Atlan workspace.

Enable Lakehouse​

To start using the Lakehouse, you first need to enable it through the Atlan Marketplace.

  1. In your Atlan workspace, navigate to the Workflow page in the left-hand navigation bar.

  2. In the Workflow page, open the Marketplace tab.

  3. In the Marketplace tab, select the Atlan Lakehouse tile. You can also use the search bar to find the Atlan Lakehouse tile.

  4. Click Enable. When prompted, enter your tenant name in the format company.atlan.com, and then click Confirm. You can receive an email notification once Lakehouse is successfully enabled.

  5. Check the status shown in the Atlan UI:

    • Initializing: Infrastructure is being provisioned. Expected duration: 15–30 minutes. If this persists beyond 60 minutes, open a support ticket. Don't restart or reinstall the workflow.
    • Hydrating lakehouse: Catalog data is being extracted into Iceberg tables. Expected duration: 1–8 hours. If this persists beyond 24 hours, open a support ticket.

  6. Once the status shows Ready, setup is complete and your OAuth credentials are available. You can now connect a query engine to start querying your metadata.

Next steps​

After the Lakehouse is enabled for your workspace, connect an Iceberg REST-compatible client to start querying your metadata.

  1. Get connection details: In the Marketplace, open the Atlan Lakehouse tile and select View connection details. Note the catalog URI, catalog name, and OAuth credentials to configure your client.

  2. Connect your query engine: See Connect query engine for guides on connecting Snowflake, Amazon Athena, PySpark, Google BigQuery, and Databricks to the Lakehouse.

  3. Query with AI agents: To query Lakehouse using natural language, install the atlan-lakehouse agent skill for Claude Code or other AI agents.