Translation of ER Model to Multidimensional Model for Data Warehouse - An Automated Approach

Abstract

Multidimensional data model has been considered by the database community as a model to support the application of data warehouse. While this model has been the de-facto of successful data warehouse implementation, little has been said on how to carry out the data warehouse conceptual design. This paper, however, presents an approach and a tool for the automatic conceptual data warehouse design. The approach uses an existing ER model describes by using an ER specification language, whereby the model is progressively translated and extended to include the dimensional functionality necessary in data warehousing. The output of the translation model is a basic multidimensional construct such as facts, measures, dimensions, and dimension hierarchies. Although there is no universally agreed method on the conceptual design of data warehouse, the approach and tool presented in this paper has at least demonstrated the practical exploitation of ER model as a basis for such a design. The paper concludes with a discussion of areas of future research development.