Looking for some advice / confirmation of my understanding regarding SSAS Cube and underlying relational data warehouse partitioning.
The SSAS Multidimensional Performance and Operations Guide (http://download.microsoft.com/download/0/F/B/0FBFAA46-2BFD-478F-8E56-7BF3C672DF9D/Microsoft%20SQL%20Server%20Analysis%20Services%20Multidimensional%20Performance%20and%20Operations%20Guide.pdf discusses the use of Matrix Partitioning for large Cubes. This is where Cube Partitions are created on date (typically) and some other key to enable parallel partition processing.
Does this mean that the underlying relational data should also be partitioned in a 'matrix' fashion? I know that one can only partition on a single column in the SQL Server Database Engine so I would need to create a persisted, computed column consisting of the Date Key and whatever other key I chose.
Is my understanding correct? Has anyone implemented such a strategy?
Thanks and regards