Metadata Database vs. Data Warehouse
Understand the difference between CARTO's metadata database and the data warehouse
Last updated
Was this helpful?
Understand the difference between CARTO's metadata database and the data warehouse
CARTO Self-Hosted involves two distinct database systems that are often confused. This page clarifies their roles.
The metadata database is a PostgreSQL instance that CARTO Self-Hosted uses internally to store platform state: maps, workflows, connections, users, and organization settings. You provision and manage this database as part of your CARTO infrastructure.
Managed by: You (customer)
Technology: PostgreSQL
Contains: Platform metadata (not your spatial data)
Size: Typically small — scales with the number of CARTO users and maps
The data warehouse is the external database where your spatial data lives. CARTO connects to it to run analyses, visualize data, and power maps. CARTO does not manage this database; it connects to it using credentials you provide.
Managed by: You (customer) or your cloud provider
Technology: BigQuery, Snowflake, Redshift, PostgreSQL, Databricks, etc.
Contains: Your spatial datasets and analysis results
Size: Scales with your data volume
Purpose
Store CARTO platform state
Store and query your spatial data
Technology
PostgreSQL
BigQuery, Snowflake, Redshift, etc.
Managed by
Customer (as part of CARTO infra)
Customer (external system)
Required
Yes, always
Yes, at least one connection required
CARTO accesses it
Internally, always
Via connections you configure
Size
Small (metadata only)
Large (your data)
The metadata database is always accessed by CARTO internally. The data warehouse is accessed on behalf of users when they open maps, run workflows, or execute queries.
How CARTO uses the Metadata Database — what is stored and why
Metadata Database prerequisites — supported versions, managed service options, and checklist
Plan Your Deployment — prerequisites, sizing, and deployment method selection
Last updated
Was this helpful?
Was this helpful?