Hello.

I have a fact table with the following design:

Fact_Table
dim_key
dim_key
dim_key
measure1
measure2
measure3

This table will contain about 1.2 million rows. Each row will contain
information about 1 insurance claim. I need to be able to determine claim
number for each row on the table. Claim number is a character field.

I tried to add claim number as an additional measure to the fact table but
when I did the aggregations for the other measures where incorrect.

I also created a dimension table as follows:

dim_policy
Policy char
Claimnumber char

However there are over 330,000 policies containing over 477,000 claims. I
used the "automatic group creation" setting when building the dimension so
Analysis Services would build the cube successfully. However when I tried to
use the dimension in the cube it took forever to drill down into individual
claims.

What is the best way to design this so I can quickly determine the claim
number of a specific row on the fact table?

Thanks.

Without doubt that it takes very long time to display too many members.

If Policy dimension is just for determining calim number, how about using
drillthrough instead?

Ohjoo Kwon


"Jeff" <Jeff (AT) discussions (DOT) microsoft.com> wrote