When you discover table relationships, the ILM engine stores the suggested unique columns in a profiling suggestion table in the ILM repository. You can view the suggested unique columns after you discover table relationships.
View the unique columns when you want to import unique keys that are not identified in table relationships. For example, you may want to maintain unique keys in the ILM repository that are not part of a primary key and foreign key relationship. When you import suggested table relationships, the ILM engine automatically creates unique keys from the suggested table relationships.
All of the discovery tools store the results of the discovery in the same suggestion table. The results may not display all of the table columns. For discovery from Data Quality, the results are filtered based on the conformance criteria that you configured when you generated the profile. For example, if you configured 90% as the conformance criteria, then you only see columns that contain 90% and higher uniqueness.
You access the suggested unique columns from the same application version or application module that you ran the discovery from. Review the list of suggested unique columns. Then, select which suggestions you want to import to the source metadata in the ILM repository. When you import the suggestions, the ILM engine adds the unique constraint type.
The suggested unique column results show the following columns:
Schema
Schema of the table that contains unique keys.
Table
Table that contains unique keys.
Column
Column that contains the unique key.
Uniqueness Percentage
Percentage of uniqueness for the column.
Available for discovery from Data Quality and discovery from Data Quality profile results.