The DI data set processing model takes source data through to data that is easily consumed, such as a measure on a dashboard. Think of this processing model as a pipeline of data with a series of refining steps along the way. This processing model is divided into “stages”, each of which serves a particular purpose:
- The Extract stage procures data from one, or typically more, sources systems.
- The Standardize stage converts this raw information into a standard form.
- The Transform stage applies business rules to standard data and incorporates reference data.
- The Organize stage is responsible for modeling the data into a form that is useful to applications, and ultimately users.
- The Presentation stage provides services to applications for delivering information to the user community.
See also Spectre measures Options.