In a typical scenario, the
hierarchical attribute representing the most detailed level will be the primary key
of the respective dimension. Each such attribute will have a corresponding foreign
key in the fact table.
In Figure 1(a) we depict an example schema of a simplified data warehouse. The
data warehouse stores sales transactions recorded per item, store, customer, and
Advanced Ad Hoc Star Query Processing
Copyright ?© 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
date. It contains one fact table SALES_FACT, which is defined over the dimensions:
PRODUCT, CUSTOMER, DATE, and LOCATION with the obvious meanings. The
single measure of SALES_FACT is sales representing the sales value for an item
bought by a customer at a store on a specific day. The dimension hierarchies are
depicted in Figure 1(b).
The dimension DATE is organized in three levels: Day-Month-Year. Hence, it has
three hierarchical attributes (Day, Month, Year). The PRODUCT dimension is organized
into three levels (Item-Class-Category) with three hierarchical attributes
and one feature attribute (Brand). The dimension CUSTOMER is organized in only
two levels (Customer-Profession) with two hierarchical attributes and two feature
attributes (Name, Address). The LOCATION dimension is organized into three levels:
store-area-region, meaning that stores are grouped into geographical areas and
the areas are grouped into regions.
Pages:
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280