key> AND
LP AND LP AND ??¦ AND LPk AND
GROUP BY
HAVING
ORDER BY
Figure 3. The ad hoc star query template
Advanced Ad Hoc Star Query Processing 4
Copyright ?© 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
the months of the year 1999 and for products that belong to the category ???air condition.???
Figure 4 shows the corresponding SQL expression of this query. One can
easily see that the query is an instance of the query template of Figure 3.
Star.Query.Processing.
Methods.of.Ad-Hoc.Star.Query.Processing.
The most well known technique for star-query processing is based on a star-join
via bitmap index intersections. Star join processing has been studied extensively
and specific solutions have been implemented in commercial products. See also
Chaudhuri and Dayal (1997) for an overview.
The standard query processing algorithm for a star join over n dimensions first evaluates
the predicates on the dimension tables, either on a normalized (snowflake) or
a denormalized (star) schema, resulting in a set Ri of ni tuples of dimension Di (1 ?‰¤
i ?‰¤ n). It then builds a Cartesian product of the dimension result tuples (R1 ?— R2 ?—
??¦ ?— Rn).
Pages:
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288