SSAS Discretization Buckets - Grouping Attribute Members in a dimension
User Rating: / 0
PoorBest 
Written by Chris Schmidt   
Thursday, 07 February 2013 21:57

Did you know you can group data through SSAS inherently? Sometimes in business the user doesn't want to see all the single values, but groups of those values. A customer's yearly income or age, for example. The business doesn't necessarily care that someone is 32 years old, but they do care that they are between the 25-34 year old age range. Or someone that makes $55,000 a year is within the range of $50,000-$75,000 a year, and households with that income are some of their best customers. One common answer to this problem is to create an additional column in your dimension and handle the grouping in your ETL, but let's say you either can't modify the ETL or would just prefer not to. Additionally, that's a lot of extra work for something that's already built in. There is a property at the attribute level within dimensions in Analysis Services that allow you to group your attribute members at run time. It's called "Discretization", and it's very easy to set up.

Read more...

 

 

Pyramid Analytics