SEARCH
0-9 A B C D E F G H I J K L M N O P Q R S T U V W X Y Z
Prev | Current Page 108 | Next

Robert Wrembel and Christian Koncilia

"Data Warehouses and Olap: Concepts, Architectures and Solutions"

Notice that the two dimensions have the same hierarchy
domain. However, in the new dimension the products are broken up and stored
in different categories for each of the four structures of Figure 3. Figure 10 shows an
unbalanced homogeneous version for the heterogeneous dimension of Figure 4.
The model provides flexibility to define a different base fact table for each bottom
category, a single base fact table for all the categories, or both. If we consider different
fact tables, the hierarchy paths of the former are consistent. Now the cube
view CV[Department] can be correctly computed by the rollup:
ROLLUP Category TO Department
If we consider cube views defined over a single base fact table that contains all the
elements, the hierarchy paths are not consistent. In this case we need ROLLUP
operations that combine categories, as we will see in the next section.
Kimball (1996) proposes a similar solution to deal with heterogeneity, but in his
approach the homogeneous structures mixed are placed in separate dimensions.
Figure 10. An unbalanced dimension that models the Product dimension of Figure 4:
(a) hierarchy schema; (b) hierarchy domain
(a) (b)
Handling Structural Heterogeneity in OLAP 4
Copyright ?© 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.


Pages:
96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120