Entity domain graphs

PolicyCenter uses entity domain graphs to define the aggregate cluster of associated objects that it treats as a single unit for purposes of data destruction. Each aggregate cluster has a root and a boundary.

• The root is a single specific entity that the aggregate cluster contains. The root entity is the main entity in the graph. A root entity is application-specific and must implement DestructionRootPinnable. Pinnable root is another term for one of these root entities.

  In PolicyCenter, the root entities are Policy, Account, PolicyTerm, PolicyPeriod, PolicyTerm, and Contact.

• The boundary defines what is inside the aggregate cluster of objects. It identifies all the entities that are part of the graph.

  In PolicyCenter, the boundary defines the entities that relate to an Account or Policy, such as jobs and effective dated fields, or to a Contact,such as an address.

The unit of work for personal data destruction is a single instance of the domain graph, such as an Account and all its associated entities.

To enforce the boundaries of the domain graph, all objects participating in the destruction process must implement the following delegate:

• Extractable – All entities in the domain graph must implement the Extractable delegate.

You cannot define a new entity domain graph for use in personal data destruction. However, you can configure an existing entity domain graph in the same way that you configure the archiving domain graph. For example, use the archivingOwner data model attribute and the CrossDomainPublicID data model tag. If personal data destruction is enabled, then at server startup, if there are validation issues with any entity domain graph, those issues are logged and the server fails to start. This process is similar to how archiving domain graph validation works.

See also

• The archive domain graph
• <delegate> elements and related data object types