Towards Model-Driven Evolution of Data Warehouses

Abstract

Data warehouse systems represent centralized data collections used for the purposes of data analysis and decision support. Development and maintenance of extensive data warehouse systems require appropriate support through methods and tools. Therefore, we introduce the project Computer-Aided Warehouse Engineering (CAWE). It is a model-driven approach to the engineering of data warehouse systems. Especially the process of data warehouse evolution, i.e. the maintenance of the core data storage component, the data warehouse in narrow sense, is a tedious task since the change of data structures implies the challenge of ensuring integrity and history of corporate data. The paper at hand provides research in progress and suggests the adoption of a multidimensional algebraic formalism to the model-driven development paradigm.