The extract-transform-load architectural pattern, as the name suggests, has the following three stages:
- Extract data from external sources
- Transform the extracted data as needed
- Load the newly transformed data into a repository specific to the current system
The most common uses of the extract-transform-load pattern is in business intelligence, data warehousing, knowledge management systems, and customer relationship management systems.
We will look at the extract, transform, and load steps as they apply to the similarly-named architectural pattern in the next sections. We will use the example of a sales tool that mines various sources of customer leads, transforms that data into a common form, and then loads the data into a central database.
