Catalog
A Catalog is a collection of Databases. When a Nodegroup is linked to a Catalog, it gains access to all Databases within that Catalog. This setup enables enterprise-level data development collaboration and data sharing scenarios—supporting multiple teams and multiple Nodegroups sharing multiple Databases through a single Catalog.
Catalog Usage
-
One Data Cloud can host multiple Catalogs. Related Databases can be grouped under a single Catalog, making all Databases within that Catalog accessible to any Nodegroup bound to it. Databases not intended for sharing can be isolated under different Catalogs.
-
All Nodegroup instances bound to the same Catalog must use the same version to ensure metadata compatibility. When upgrading versions, all Nodegroup instances under the Catalog need to be upgraded simultaneously.
-
During Nodegroup initialization, users can either create a new Catalog or choose to bind to an existing Catalog.
-
Once a Nodegroup is bound to a Catalog, it cannot be bound to another Catalog.
Create a Catalog
When creating a new Nodegroup, select "New" to create a Catalog. The new Catalog will be initialized with system default Databases: postgres, template0, and template1 by default.
Bind to a Catalog
When creating a new Nodegroup, select an "Existing" Catalog to bind. After linking, the Nodegroup cannot be unbound from the Catalog—if necessary, delete the Nodegroup and then recreate it.
View Catalogs
Under the "Data" menu, view existing Catalogs. Within each Catalog, Databases can be viewed and managed.
Delete a Catalog
To delete a Catalog, first remove all user-created Databases in the Catalog. System default Databases can be ignored.