blogs.conchango.com

It is generally considered data warehousing best practice to split your facts into different tables if the granularity/dimensionality is different.  This is a very common question in data warehouse design.  Consider the following data model.   The company traditionally sells its products through sales people visiting customer ...

I’m approaching the end of an enjoyable data warehousing project and we have collected some size/volume statistics around our Analysis Services 2005 cube.  We have: * 48 Dimensions.  This includes role-playing dimensions.* 751 Attributes and user hierarchies (!).  So if a dimension has 5 attributes and 2 user hierarchies, ...

One of the key goals of Analysis Services 2005 is to “bring together the best of both relational and OLAP reporting into a single Unified Dimensional Model”.  This is a very ambitious goal and most would agree that it would be difficult to fully achieve.  However, Analysis Services 2005 certainly does take huge strides in ...

I once wrote an add-on for the SQL Server Accelerator for Business Intelligence to provide unknown-member functionality.  I was therefore naturally interested when I heard that it comes out-of-the-box with Analysis Services 2005. The idea with the unknown member is that, for each level of a hierarchy (or for each attribute in Analysis ...