Question on partitions on cubes RRS feed

  • Question

  • Hi,all experts here,

    Thank you very much for your kind attention.

    I am having quesions about the partitions on cube. Will the partitions actually improve the query to a cube (after it is divided into different partitions?), or it just decrease the time cube administrators used to process the cube?(in this case, only needed to process the partition independently when update happens).

    Also, how about the aggregations? Will the UMD still generate the aggregates on the whole cube by adding all partitions? Or aggregates only for each partition?

    Also, when creating partition, how do we actually decide the level of performance for the aggregations?

    Hope my question is clear.

    I am looking forward to hearing from you.

    Thanks a lot for your help.

    With best regards,

    Yours sincerely,


    Tuesday, April 24, 2007 3:20 PM