Be Careful Partitioning Analysis Services (Tabular)
User Rating: / 1
Written by Bob Duffy   
Wednesday, 26 February 2014 22:14

There are generally three reasons for partitioning an Analysis Services model:

  1. To improve processing time by running partitioning in parallel
  2. To improve processing time by allowing for incremental processing of some sort
  3. To improve query time where most queries only need a smaller slice of data - say a week or month.
  4. /ol>




XL Cubed