DataWarehouse Data Model - financial Serivces

[mishka]

Hello,

we are creating a new DW at our firm. I have several questions. The first one is In regards to fact tables. I am starting to build this table out and noticing that the number of measure will be extremly high. I was wondering if anyone has any ideas/recommendation on the subject.

We are interested in using a BI tool in the future. Most likely Analysis Services.

Thank you in advance.

m

[Jeff Smith]

It depends. If the measures are populated for each row, then maybe a lot of measures is OK. If few rows have values for every measure and if most of the measures are blank or 0 for the rows, well then maybe you add a Measure Dimension to the fact and have just one measure. But if a lot of rows have a lot of measures, then the number of rows in your fact table will grow greatly with a measure dimension.

[mishka]

Hi,

There are about 15 of the 60 measure colunms that are like buckets and will be 0 for most of the attributes. However, In those cases we would want to see the 0 value. Any advise you have will be appreciated.

thank you,
m

[Sponsored content]