The advisability of using packages in data warehouse design

Abstract

Data warehouses are large data repositories integrating data from several sources that support decision making. Although, traditionally, data warehouses have been designed using the ‘well-known’ star schema, some design methodologies have come into existence in recent times. These new methodologies have not only focused on logical design: they also propose performing a conceptual modeling using UML. At present, it is widely accepted that modeling using packages simplifies the management and understanding of the designs.Until now, however, this statement has not been empirically proved in the data warehouse field. In this paper, we present an empirical study whose aim is to check whether using packages in designing data warehouses makes them more understandable.