6.2. Loading Aggregate Dimensions
The process of loading an aggregate dimension table is very similar to that of loading a base dimension table, as described in the previous chapter. An incremental load for the aggregate dimension must extract and prepare records, identify new records and type 1 and type 2 changes, and manage surrogate keys.
As you have seen, many aspects of the load are simplified because the aggregate is sourced from the base schema. The elements that are simplified include identification of changed data and assembly of dimension attributes.
Other aspects can become a serious challenge. Processing a type 1 change to an attribute that is not dependent on any other attribute in the dimension presents numerous difficulties for aggregates.
It is also easy to make the dimension load more complicated than it needs to be. As you will see, the common practice of mapping dimension keys to aggregate dimension keys is not necessary for fact loads.
6.2.1. Requirements for the Aggregate Dimension Load Process
As you have seen, an aggregate dimension table is a dimension table itself. The process that loads an aggregate dimension, then, is subject to the same top-level requirements. As originally presented in Chapter 5, they are:
Extraction of the necessary source data
Assembly of the dimensional attributes
Warehouse key management
Processing of new records
Processing of type 1 changes
Processing of type 2 changes
There is nothing special about an aggregate dimension table that ...
Get Mastering Data Warehouse Aggregates: Solutions for Star Schema Performance now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
