In OLAP, queries are essentially views that aggregate raw facts to a granularity
specified by a list of categories, called granularities, selected from a list of dimensions.
The central technique for speeding up OLAP query processing is to materialize
(precompute) some aggregate views and use them for the derivation of other
aggregate views, using operations called roll-up operations. Aggregate navigation, a
central technique in OLAP, is the process of finding and testing correct derivations
of aggregate views from other precomputed aggregate views at lower granularities.
Since the cost of computing a cube view is essentially linear in the number of facts
accessed in the computation, this technique can speed up this processing by a factor
which can be up to several orders of magnitude in real scenarios (Kimball, 1995).
The central problem caused by heterogeneity is the obsolescence of the framework
Handling Structural Heterogeneity in OLAP
Copyright ?© 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
of OLAP aggregate navigation that underlies OLAP systems, which yields the inef-
ficiency of computing the aggregate views always from the raw facts.
Content.
In this chapter, we show how to incorporate structural heterogeneity in OLAP data
models.
Pages:
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97