Defining an in-database aggregate with a parent-child dimension

An in-database aggregate can contain a parent-child dimension. Because the dimension does not have hierarchy levels, you create the relationships by mapping a single column in the aggregate table to the child key in the parent-child dimension.

The in-database aggregate can also contain dimensions with level-based hierarchies. For more information about adding these dimensions, see Defining an in-database aggregate manually.

Procedure

  1. Select the dynamic cube in which you want to define an in-database aggregate from the Project Explorer tree.
  2. Click New In-database Aggregate New In-Database Aggregate icon.
  3. Select the measures and parent-child dimension to include in the in-database aggregate, then click OK.

    An in-database aggregate is created, which also appears under the In-Database Aggregates folder in the Project Explorer tree.

    Now map a single column in the aggregate table to the child key in the parent-child dimension.

  4. In the Project Explorer tree, double-click the in-database aggregate in the In-Database Aggregates folder.

    The In-Database Aggregate editor is shown.

  5. Select the parent-child dimension, then select the I want to remap the columns for this dimension, as they are included in my aggregate check box.
  6. Click the Key Mappings tab.
  7. Drag a column from the required aggregate table in the Data Source Explorer to the Mapping field for the child key.

    Next, you must map measures in the in-database aggregate to columns in the aggregate table.

  8. In the In-Database Aggregate editor, click Measures measures icon.

    The Measures editor is shown.

  9. Map each measure to a column in the aggregate table by dragging a column from the required aggregate table in the Data Source Explorer onto the Mapping field.

Results

The in-database aggregate is complete. You now test the validity of the in-database aggregate. For more information, see Validate a project and individual objects.