Dependent Data Marts
If you need to develop one or more physical data marts in the Teradata environment, you should strongly consider configuring them as dependent data marts. Dependent data marts can be built in one of two ways: either where a user can access both the data mart and the complete data warehouse, depending on need, or where access is limited exclusively to the data mart. The latter approach is deprecated and the type of data mart it produces is sometimes referred to as a data junkyard.
In the data junkyard, all data begins with a common source (in this analogy, “cars”), but they are scrapped, rearranged, and generally junked to get some common parts that the yard operator believes are useful to his customers. The parts collection in the junkyard relates more to what has been useful in the past: previous supply and demand determines what the user can access.
Continuing the analogy, you, as a user, visit the junkyard and search through the various wrecks you encounter in hopes of finding the part you need. To find your part (to answer your question), you will probably need to scavenge through several different junkyards.
The approach results in a decision support environment molded, and compromised, from a specific, well known set of questions and responses rather than around your ever-changing business needs.