Retrieving Multidimensional Objects

During the design of a data warehouse, you will need to identify which of your tables and views will represent facts, and which dimensions. In this section, we will refer to both tables and views as being either:

• A fact table – which stores variable numerical values related to aspects of a business (for example, sales, revenue, budget). These are usually the values you want to obtain when you carry out a decision support investigation.

• A dimension table – which stores data related to the axis of investigation of a fact (for example, geography, time, product). A dimension table should be connected to a central fact table.

You can specify your tables and views individually as facts or dimensions, or you can use Multidimensional Objects Retrieval Wizard to perform this task automatically, based on the references connecting them:

• Child tables or views become Fact tables or views

• Parent tables or views become Dimension tables or views

The new type is indicated in the Dimensional Type field in the object's property sheet, and a type icon is displayed in the upper left corner of its symbol:

Fact table	Dimension table

1. Select Tools > Multidimension > RetrieveMultidimensional Objects to open the Multidimensional Objects Retrieval Wizard.
2. Specify the objects to be retrieved. By default both Facts and Dimensions will be retrieved.
   Note: If you are working with Sybase AS IQ v12.0 or higher, you can also select to automatically rebuild join indexes after retrieving multidimensional objects. For more information, see IQ Join Indexes.
3. [optional] Click the Selection tab to specify which tables to retrieve multidimensional objects from.
4. Click OK to retrieve the multidimensional objects..

   The selected tables are assigned a multidimensional type.

• Rebuilding Cubes
  Once the fact and dimension tables of the data warehouse schema are designed, you can use this information to build the multidimensional cubes. The Rebuild Cubes Wizard transforms fact tables or views into cubes, and dimension tables or views into dimensions. You can then design the cubes taking into account the different analysis axes of the dimensions. These cubes will serve to generate the text files used to create and populate the OLAP engine.