Dynamically build aggregations RRS feed

  • Question

  • I have an SSIS process that dynamically builds partitions as new data becomes available in the source system.  One thing I am trying to figure out is how to emmulate the "build aggregation" wizard via SSIS.  For example, lets say I have a new partition and want to have a higher number of aggregations versus an older one. If I change the aggregation percentage and run PROCESS INDEX, will that take care of it?  Or lets say I define new aggregations (higher %), dont I have to tell it to point to the next aggregation design?  So in short, can SSIS mimick the aggregtion wizard?

    One of the things I am struggling with is to understand how the aggregation design fits into all of this.  This is probably a second question, but when you dynamically build partitions how do you set the correct aggregation design?  Can you create one agg design and assign it to all partitions?  The way it was described to me is that the aggregation design that is set for the partition is the one that the cube will use to look for an aggregation.  Wouldnt this change from partition to partition? 

    In my SSIS package, I set the aggregation design to a default aggregation.  So my thought was that if you built aggregations dynamically via SSIS, you could ensure you were using the correct aggregation design.

    The example MSFT touts is the one from Project REAL, but that was a ROLAP partition scheme that did not deal with aggregation designs. 

    Any help would be very much appreciated.

    Monday, January 29, 2007 10:32 PM