Can someone comment on (or suggest another approach) based on the following statement?
| A search predicate that spans several dimensions resolves to a | |
| set of tuples (pairs, triplets etc of members of the referenced | |
| dimensions). This set may be huge because the set of all tuples | |
| is the Cartesian product of members of these dimensions. This is | |
| one reason why it has not been implemented. If the dimensions are | |
| large it may be better to extract a large result set from a cube | |
| than generate a humongous MDX statement that may take longer to | |
| parse and interpret or may even exceed DBMS limits. |