Introducing the Schema Generation Wizard
When you design your dimensions and cubes by using the top-down method in Business Intelligence Development Studio, you create dimension and cube definitions in a Microsoft SQL Server Analysis Services project and then use the Schema Generation Wizard to generate a data source view, a data source, and the underlying relational database schema that supports these OLAP objects. This relational database is referred to as the subject area database.
After the Schema Generation Wizard has generated the underlying objects based on the design of your dimensions and cubes in an Analysis Services instance or in an Analysis Services project, you can change the design of the dimensions and cubes, and then rerun the Schema Generation Wizard to regenerate the underlying objects based on the modified design. When the underlying objects are regenerated, the Schema Generation Wizard incorporates the changes into the underlying objects and, as much as is possible, preserves the data contained in the underlying databases.
After the Schema Generation Wizard generates an iteration of the relational schema, you can modify the dimensions and cubes upon which the relational schema was based, and then use the Schema Generation Wizard to regenerate the subject area database schema. For example, after you initially generate the subject area database schema, you can add or modify cubes, dimensions, attributes, and measures, and then use the Schema Generation Wizard to regenerate the previously-generated subject area database.
When the underlying objects are regenerated, you can choose to have the Schema Generation Wizard preserve the existing data in the tables being regenerated in the subject area database, so that you do not need to reload or reenter the data between iterations. However, under some conditions, not all data can be preserved. For example, some data must be dropped if you delete a dimension or an attribute that contains data. If the Schema Generation Wizard must drop some data because of a schema change, you receive a warning before the data is dropped and can then cancel the regeneration.
As a general rule, any change that you make to an object that was originally generated by the Schema Generation Wizard is overwritten when the Schema Generation Wizard subsequently regenerates that object. The primary exception to this rule is when you add columns to a table that the Schema Generation Wizard generated. In this case, the Schema Generation Wizard preserves the columns that you added to the table, as well as the data in these columns.
For more information about incremental generation, see Understanding Incremental Generation.