Chapter 4
This chapter describes the process of taking the conceptual design you have discovered through the business analysis process and deriving a database design that can be implemented as a dimensional schema.
Creating a dimensional schema is very different from creating a relational schema: A relational schema usually has a firm set of specifications that must be satisfied for a database design to be successful for example, "This order entry system must store detailed information about customers and orders." As long as the system meets these requirements, it can be considered a success.
Typically, the specifications for a dimensional design are not firm. Furthermore, a well-designed data warehouse database will often support capabilities that were not previously anticipated. For example, a data warehouse about customers may provide additional business intelligence about shipping company efficiency.
While creating a dimensional schema differs from designing a relational schema, the two processes share several common design practices that are used to construct a useful dimensional design. This chapter discusses those practices and some of the issues that arise when a dimensional design must evolve to handle changes.
To complete the exercises in this chapter, you must