Conceptual multidimensional modeling for data warehouses: A survey

Abstract

Conceptual multidimensional modeling aims at providing high level of abstraction to describe the data warehouse process and architecture, independent of implementation issues. It is widely accepted as one of the major parts of overall data warehouse development process. In the last several years, there has been a lot of work devoted to conceptual multidimensional modeling for data warehouses. This paper presents a survey of various proposed conceptual multidimensional models for core as well as advanced features supported. Hereafter, a comparison of the models is done based on the criteria broadly categorized as: general aspects, relationship aspects and implementation aspects, showing the evolution in this area. General aspects involve basic features of the multidimensional model. The relationships among various constructs used in the multidimensional model are referred under relationship criteria. The implementation criteria relate aspects such as mathematical formalism and guidelines to define complex multidimensional structures.