Oracle Tips  by Burleson Consulting

The Data Warehouse Development Life Cycle

The Role Of Functional Decomposition As you can see, each behavior within the database now corresponds with a single database update operation. The insert on the line_item record is associated with the add_item behavior. This one-to-one partitioning is only important for database updates, as select operations may still access many records within the database.

There is still a great deal of controversy about the best way to approach database analysis for data warehouse systems. Architecturally, some theoreticians state that the relational model is better suited for use in an online transaction processing environment and multidimensional architectures are better suited to data warehouses. Oracle Corporation has stated that they will incorporate a user-defined datatype in their future release of Oracle8. Some argue that a database must be able to have datatypes that are lists rather than finite values, and some databases, such as UniSQL, allow for single datatypes (fields) to contain lists of values or even another table.

Developers must remember that the main difference between traditional systems analysis and data warehouse analysis is the focus on the data sources. Because a data warehouse will never have data coming from an online data input screen and all of the end user interfaces will be ad hoc in nature, the data identification, loading, and query processing is of foremost importance.

If you like Oracle tuning, you may enjoy the book Oracle Tuning: The Definitive Reference , with over 900 pages of BC's favorite tuning tips & scripts.  You can buy it directly from the publisher and save 30%, and get instant access to the code depot of Oracle tuning scripts.