Siebel Analytics Server Administration Guide > Creating and Administering the Physical Layer in a Repository >

Creating and Administering Physical Tables

For all data sources, you can define general properties, columns, a primary key, and foreign keys. However, because you cannot import tables for multidimensional data sources, you must create each physical table cubes and define its properties.

Physical Tables for Multidimensional Data Sources

Importing multidimensional physical schema is not supported at this time. For multidimensional data sources, you need to manually create all elements of a physical table cube. This includes columns, a primary key, foreign keys, and hierarchies.

Each cube from a multidimensional data source is modelled as a physical cube table, a type of physical table. It has all the capabilities of a table such as columns (physical cube columns, not regular columns) and keys (optional) and foreign keys (optional). It also has cube-specific metadata such as hierarchies and levels.

In the Physical layer, a physical cube table looks like a regular table but will have a different icon. For more information, see Icons and Symbols in the Administration Tool.

NOTE:  It is strongly recommended that you build each cube one hierarchy at a time and test each one before building another. For example, create the time hierarchy and a measure and test it. When it is correct, create the geography hierarchy and test it. This will help make sure you have set up each cube correctly and make it easier to identify any setup errors.

For supported data sources (not multidimensional), you can import the physical schema. If the keys are defined in the database and the table definitions are imported with their columns and keys, the table properties are configured automatically.

To create a Physical table or a Physical cube table and any necessary properties, perform the tasks in the following list:


 Siebel Analytics Server Administration Guide
 Published: 11 March 2004