【Microsoft Purview】How build hierarchy relationship for Azure Data Lake Storage

To build a hierarchy relationship for Azure Data Lake Storage entities in Microsoft Purview, you will need to define the relationships between the various entity types such as azure_datalake_gen1_account, azure_datalake_gen1_resource_set, azure_datalake_gen1_object, and azure_datalake_gen1_path.

While the built-in entity types exist, the relationship definitions may not be explicitly provided in the relationshipAttributeDefs property for these entities. To establish these relationships, you can create custom relationship types that define how these entities interact with one another.

1. Define Relationship Types: You can create custom relationship types that specify the nature of the relationship (e.g., composition, aggregation, association) between the entities.
2. Use the IsHierarchical Property: If you are implementing a self-referential one-to-many relationship, set the IsHierarchical property to true to enable hierarchy viewing and querying capabilities.
3. Implement Relationships: After defining the relationships, you can implement them in your data model, allowing you to query and visualize the hierarchical relationships effectively.

For more detailed guidance on creating custom types and relationships, refer to the Microsoft Purview documentation.

Hi Mofei Zhuang,
Welcome to the Microsoft Q&A and thank you for posting your questions here.
Microsoft Purview’s built-in ADLS Gen1 types (azure_datalake_gen1_resource_set, azure_datalake_gen1_path, azure_datalake_gen1_object) do not define parent-child relationships in relationshipAttributeDefs. Hierarchy is inferred from qualifiedName (e.g., account/resourceSet/path/object). If explicit links are required, create custom Atlas relationship types and attach them via Purview REST API:

Steps:

1. Define a relationship type using POST /api/atlas/v2/types/typedefs with relationshipDefs.
2. Link entities using POST /api/atlas/v2/relationship with GUIDs of parent and child.
3. Use qualifiedName patterns to derive hierarchy programmatically.

Please let us know in the comments if you have any further questions or would like to continue the discussion. We’ll be happy to assist.