Catalogs
You can use a catalog to easily find and share your data and other assets. A catalog is like a private community for your organization. It’s a way to organize resources for many data science projects: data assets, models, other types of assets, and the users who need to use the assets.
Requirements
: IBM watsonx.data intelligence or IBM watsonx.ai Studio
Restrictions
If you have watsonx.ai Studio but not IBM watsonx.data intelligence, then there is a limit of one catalog. The catalog is limited to no more than 50 assets and no more than 50 collaborators.
Watch this short video to learn about IBM watsonx.data intelligence.
This video provides a visual method to learn the concepts and tasks in this documentation.
Accessing catalogs
To access a catalog, you must be added as a collaborator. To see your catalogs, go to the navigation menu and select Catalogs > View all catalogs. You can sort catalogs by name or date created.
Catalog components
A catalog consists of the following components:
- A metadata repository that acts as an index for data and other assets.
- A list of collaborators who need to access the assets.
- An catalog storage instance to store assets that are copied into the catalog.
Catalog roles and associated tasks
How you can use a catalog depends on your role within the catalog. You might have different roles in different catalogs. To see your role, go to the Access control tab in the catalog and find your name.
| Role | Tasks |
|---|---|
| Viewer | • Find and view assets • Copy assets to projects to work with them |
| Editor | • All Viewer tasks • Add assets to the catalog • Add catalog assets to a project • Modify assets • Remove assets |
| Admin | • All Editor tasks • Manage collaborators • Change catalog settings • Delete the catalog |
Catalogs for data governance
You can use a catalog for data governance and AI governance in these ways:
- You can govern data with data protection rules that deny access to data or mask data in a catalog depending on the identity of the user and other criteria.
- You can share data assets with features in a catalog that acts as a feature store.
- You can track and govern AI models in a catalog that acts as a model inventory.
Catalogs and lineage
To understand where your data comes from and how it is moved or transformed, see Data lineage.