Clever Geek Handbook

Analysis service. Performance issues processing the second dimension

I am working with a Google Analytics app that can create and process DM and Cube. The user can select the dimensions / attributes / measures / hierarchies they want to see in the Cube.

My cube loading / processing thread when I first create and load a cube

1) Create DS, DSV, dimensions, cubic structures 2) Complete processing 3) Process Full cube

After that, the user can change the dimension structure (add / remove attributes) and I need to refresh and reload the cube a second time , the flow is:

1) Update dimensions, cubic structures 2) Complete processing 3) Process Full cube

I have a performance issue when I do the Complete process a second time across dimensions because it recycles related measure groups and partitions.

Any suggestion that I can change the stream structure would be appreciated

+3


source to share


1 answer


Simple but effective changes, as Microsoft recommends, always "process the data" and then "process the indexes" instead of the "full process". Link: http://msdn.microsoft.com/en-us/library/cc966525.aspx



+2


source






More articles:


All Articles