Toward propagating the evolution of data warehouse on data marts

Abstract

Modern Decision Support Systems (DSS) are composed of a Data Warehouse (DW) which stores all data necessary for decisional purposes, and a several Data Marts (DMs). A DM is a subject oriented extract of the DW data; it facilitates evaluating the performances of a business process. However, in practice, business processes may evolve as well as new ones may be created, consequently the DW model evolves in order to capture the changes. The maintenance of DMs due to the evolution of their DW is time-consuming, expensive and error-prone process. This paper studies the evolution of DM schemas in the case of a new table is added to the DW. © 2012 Springer-Verlag.