Databases Reference
In-Depth Information
13
ETL into a Time Variant
Data Wareho
use
The purpose of an Extract, Transform, and Load (ETL) application is to
bring data from the enterprise into a d ata warehouse. That may sound
rather simplistic. However, the simplicity of that statement belies the com-
plexities of ETL. In
Building and Maintaining a Data Warehouse
, Chapter
6 explains the analysis, architecture, design, development, and implemen-
tation of an ETL application. If you are reading this chapter intending to
learn all you need to know about ETL, please refer to Chapter 6 of
Building
and Maintaining a Data Warehouse
. For a data warehouse rookie, the
text of that chapter will provide the foundation information necessary
to understand and incorporate the Time Variant ETL design discussed
in this chapter. The fundamental ETL concepts included in this chapter
include Changed Data Capture (CDC) and ETL Key. The Time Variant
ETL design will build on these two concepts.
chANGeD DAtA cAPture
entities
The CDC function applies specifically to dimension tables. Remember,
dimension tables contain the entities of the enterprise. Those entities
can include the products, locations, personnel, vendors, customers,
formulas, transportation vehicles, and so on. Entities are those things
that are somehow involved in the business processes of the enterprise.
201
