Data Wrangling

DM Review, January, 2008 by Ralph Kimball

In the first two parts of this series, we got ready to start the design of an enterprise data warehouse. First we carefully took stock of all the simultaneous requirements and constraints facing the designer. Then we reminded ourselves of the appropriate boundaries between our responsibilities and those of the other business intelligence (BI) stakeholders.

In this third column, we are ready to design the first stage of the data pipeline leading from the operational sources to the final BI user interfaces. I call this stage “data wrangling” because we must lasso the data and get it under our control. Successful data wrangling includes change data capture, extraction, data staging, archiving and the first step of data warehouse compliance. Let’s examine these narrow data...

Read the full article with a Free Trial at MyWire.

Premium Content Partnership | MyWire provides an in-depth online archive library of reference works.