A typical star join is transformed then into a multidimensional range
query, which is very efficiently computed using the underlying multidimensional
data structures. The combination of the two: hierarchical clustering of data and a
multidimensional structure for accessing the fact table tuples results in a very ef-
ficient method for ad hoc star query processing.
8 Karayannidis, Tsois, & Sellis
Copyright ?© 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of
Idea Group Inc. is prohibited.
In this chapter, we discuss the processing of ad hoc star queries over hierarchically
clustered fact tables. In particular, we present a complete abstract processing plan
that covers all the necessary steps for answering such queries. This plan directly
exploits the benefits of hierarchically clustered fact tables and opens the road for
new optimization challenges. Then we proceed in realizing this abstract plan for the
case of a multidimensional storage structure that achieves hierarchical clustering,
namely the CUBE File. We continue with a discussion on star query optimization
for the presented abstract plan and present the hierarchical pregrouping transformation.
This is a very elegant transformation that exploits dimension hierarchy
semantics to speed up query processing significantly. Finally, we conclude with a
discussion on main conclusions of the presented methods, and future trends in star
query processing.
Pages:
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278