AD HOC CLASSIFICATION

Warehouse developers often see that the ability to reclassify data attributes may lead to new “facts” within a data warehouse. As in our example, Guttbaum’s has two facts: sale_amount and quantity_sold. However, the ability to reclassify according to data attribute may lead to derivations from these base facts. These derivations might include:

* The average number of items per transaction.

* Total dollars spent per product and brand name category.

* Total dollars spent, grouped by customer category.

* The correlation quotient between attribute X and attribute Y (e.g., compare the propensity of a customer to purchase beer with diapers with the propensity of all customers to purchase beer with cigarettes).

If we accept the premise that there are n! possible combinations of possible data attributes that can be combined into a classification, then how can we possibly plan for all of the possible combinations? As it turns out, when developers build an Oracle warehouse to be flexible in nature, the base architecture can be easily extended to accommodate new classifications. Building a flexible warehouse involves planning for the ability to define new aggregate summary tables and creating an application front end that allows easy query classification. Some of these flexibility techniques will be discussed in Chapter 4, Data Warehouse Design.

This is an excerpt from "High Performance Data Warehousing".

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.