January 7, 2014 at 10:12 am
Hello!
We are currently thinking about the pros and cons of moving a relational database to a multidimensional database (aka cube). The database contains approx. 200,000,000 transactional records with 15 KPI's and gets queried using about 85 dimensions (derived from master data) which are used for aggregation and for filtering as well. In a typical query, different dimensions are used to aggregate and there are several filters using these or other dimensions.
Has anyone ever worked with a cube with that many dimensions and filters and can state something about the effects?
Any help is greatly appreciated!!!
Oliver
January 10, 2014 at 11:13 am
As a rule, when you've gone past 15 dimensions in most solutions, chances are there are some opportunities for the consolidation of dimensions. I have never personally had a solution with more than 17 dimensions no matter what the size, scale, or vertical industry.
That said, Analysis services should be able to handle 200,000,000 rows of fact data if it is in a star schema.
Viewing 2 posts - 1 through 1 (of 1 total)
You must be logged in to reply to this topic. Login to reply