You can improve the performance of federated queries that read record array data if you map a separate table for each record array in the table definition.
Each table you map separately consists of a single record array definition that contains the column definitions unique to a single array instance. Any given column appears in each instance of the array. For example, if your employee table has an array that describes employee dependents, the table structure might look like this:
KEY | DEP_LAST_NAME | DEP_FIRST_NAME | DEP_GENDER | DEP_ID | DEP_DOB |
You can map a separate table for each array by running the New Table wizard in Classic Data Architect once for each array, or you can copy and edit table objects as described in the following steps.
To map a separate table for each array in a base table by using the Data Project Explorer view: