This chapter describes concepts and techniques of dimensional data modeling and shows how to build a simple dimensional data model. Implementing a Dimensional Database (XPS) shows how to use SQL to implement this dimensional data model.
A dimensional data model is harder to maintain for very large data warehouses than a relational data model. For this reason, data warehouses typically are based on a relational data model. However, a dimensional data model is particularly well-suited for building data marts (a subset of data warehouse).
The general principles of dimensional data modeling that this chapter discusses are applicable for databases that you create with Dynamic Server or Extended Parallel Server. Although no single factor determines which database server you should use to build a dimensional database, the assumption is that large, scalable warehouses are built with Extended Parallel Server, while smaller warehouses, OLTP systems, and operational systems are built with Dynamic Server.
To understand the concepts of dimensional data modeling, you should have a basic understanding of SQL and relational database theory. This chapter provides only a summary of data warehousing concepts and describes a simple dimensional data model.
Home | [ Top of Page | Previous Page | Next Page | Contents | Index ]