Epic Goal

Provide a backend module in RHDH (that extends off of the base catalog plugin) to ingest model catalog entities into the catalog in RHDH.

The backend module should make use of Entity Providers and Custom processors[1] to communicate with the Model Catalog Bridge service to fetch model/model server entities that have been retrieved from a variety of sources (currently just RHOAI MR).

The entity provider(s) should focus on:

• Fetching catalog entities on initial startup
• Deleting removed catalog entities
• Discovering new catalog entity locations that have been added by the Bridge service

The custom processor should focus on:

• Updating catalog entities as they change
• Discovering new catalog entities that are added

A current prototype plugin that largely implements the above can be found here: https://github.com/redhat-ai-dev/rhdh-plugins/tree/main/workspaces/rhdh-ai.

[1]: https://backstage.io/docs/features/software-catalog/external-integrations/#custom-entity-providers

Why is this important?

• The backend module / plugin will form the basis of how model/model server entities get ingested into the RHDH catalog
• The backend module will determine the lifecycle of the catalog entities it will discover

Scenarios

1. Plugin Workspace: Provide a development environment and basic plugin scaffolding for us to iterate on (and determine appropriate repository to host the plugin)

1. Model Catalog Entity Provider: Develop an entity provider to ingest entities into the RHDH model catalog
2. Model Catalog Custom Processor: Develop a catalog custom processor to update model catalog entities, and to discover new models, from the model catalog bridge service

Acceptance Criteria (Mandatory)

• CI - MUST be running successfully with tests automated
• Release Technical Enablement - Provide necessary release enablement details and documents.
• ...

Dependencies (internal and external)

1. ...

Previous Work (Optional):

1. …

Open questions::

1. …

Done Checklist

• Acceptance criteria are met
• Non-functional properties of the Feature have been validated (such as performance, resource, UX, security or privacy aspects)
• User Journey automation is delivered
• Support and SRE teams are provided with enough skills to support the feature in production environment