Please start any new threads on our new
site at https://forums.sqlteam.com. We've got lots of great SQL Server
experts to answer whatever question you can come up with.
Author |
Topic |
Gmare
Starting Member
6 Posts |
Posted - 2012-06-27 : 06:29:43
|
Hi all,I'm creating some dimensions in SSAS 2008 and I noticed that private dimensions can't be created. I have the need to let possible to show only members of a dimension that have a reference in fact table, and in previous versions of SSAS it was possible creating a private dimension. Example:Customer master data: 800.000 recordsFact table: data referenced to 50.000 customers available in master dataIf I create a normal dimension, during cube browsing or using client tools I see all the master data records in filter option. My need is to let possible to select only the 50.000 customers that are in fact table.Thanks |
|
gparminder
Starting Member
1 Post |
Posted - 2012-08-01 : 10:49:47
|
Hi,Though it looks like an old unanswered post, and you might already found a solution to your query, but see if following pointers can help.I have implemented the similar scenario using mdx expressions in SSAS 2008 R2 and SSAS 2012 Cube roles as per the below steps: - Connect to SSAS cube database using SSMS - Expand the database --> Roles - double click on the role you want to implement this expression - This will open Roles properties window - Go to Page 'Dimensions Data' - Select the dimension 'Customer'. Remember to select it from Cube dimension list, which could be below in the drop down list at second level - select 'Advanced tab' and from drop down attribute list select 'Customer Name' attribute - in 'Allowed Member set', enter MDX statement similar to one of the below statement (you need to update below with the correct dimension, fact measure name and hierarchy levels as per the model you have) Filter([Customer].[Customer Name], [Measures].[ Fact Count]>0) OR NonEmpty([Customer].[Customer Name].MEMBERS, [Measures].[Fact Count]) Don't forget to check 'Enable Visual Totals' before clicking ok.I hope you find this useful. |
|
|
|
|
|