Hi,
We have a reporting application, which currently reads a denormalized
detail datamart containing .5 million rows from SQL2000 relational db.
It's a regular financial transaction, containg
document#,
date,
document type,
cost center,
business unit,
account#,
currency etc etc. There are 30 such so called dimension fields. And
there is only one fact, namely the amount. The business now wants this
info in a cube. My questions are:

- is this data really 'Cubable' or should we talk the businees out of
it ?
- if it can be cubed, could someone give the best approaches to do
this ? Do we have a build a cube with 30 dimensions and a fact ? In
fact, this 30 is not all. We also have other dimensional hierarchies,
such as cost center hierarchy, business unit hierarchy etc. using
which the cube should be queried.

We are ofcourse using SQL2000 Analysis Services on W2K Advanced
Server.
Also, any pointers like white papers/books etc. on this are also
greatly appreciated.
Thanks a lot for any help...Raju.

Hi Raju,

I would recommend that you build dimensions based on the fields that your
users will want to slice and dice by. To access additional detail, you may
want to look into the Drillthrough functionality in Books Online.

Marin Bezic

--
This posting is provided "AS IS" with no warranties, and confers no rights.
Use of included script samples are subject to the terms specified at
http://www.microsoft.com/info/cpyright.htm.


"mangaraju venuturupalli" <mangaraju (AT) yahoo (DOT) com> wrote