A Meta-model Based Approach for Data Warehouses Design and Implementation

Abstract

In this paper, we present a meta-model approach for the design and implementation of Data Warehouses (DWs). The approach is made up of a set of transformation rules as a mechanism to extract multidimensional schemas from the logical description of the operational Database and a Computation Independent Model (CIM) for user requirements definition. As a result we have implemented an Eclipse based prototype that generates from the logical description of the operational Database and user requirements the multidimensional schema that best reflects users requirements. Next, for multidimensional schemas generation, we use user requirements to guide the selection that is supported by the operational Database that is most likely to satisfy users requirements. Finally, we define a tool for the data Warehouse implementation.