Discovery uses indexes on the data sources to access data efficiently and quickly. Discovery then can discover relationships between data source tables. It is not required to index and discover a data source; you might have defined Primary Key-Foreign Key relationships, or want to configure relationships manually. But to discover relationships efficiently, you need to index and discover the relationships.

This section explains important factors to consider for the indexing and relationship discovery process and describes using data sampling for large data sources.

This topic contains: