Modeling Slowly Changing Dimensions

Slowly Changing Dimensions (SCDs) are a fact of life in almost all data warehouses; we discussed how to deal with them in our relational model in Chapter 1, Designing the Data Warehouse for Analysis Services, and we'll now look at the issues that arise with each different type when building Analysis Services dimensions. In many cases, of course, different attributes on the same dimension can be subject to different types of change and we will use a combination of the techniques outlined in the next section, but it's helpful to discuss the three most common types of change separately for the sake of clarity.

Type I SCDs

Since changes in Type I SCDs involve overwriting existing values in the dimension table, no special design is needed in Analysis Services to support this. Running a Process Update on the dimension will ensure that any changes that have been made in the dimension table are reflected in your Analysis Services dimension. We'll discuss processing...