Towards a Generic Conceptual Model for Data Warehouses

Abstract

Data Warehouse systems enable managers in corporations to acquire and integrate information from heterogeneous sources and to query huge databases efficiently. Building a Data Warehouse requires focusing closely on the conceptual design phase due to the specific requirements found in the conceptual model used where we argue that the famous Entity-Relationship model is not suitable for modeling Data Warehouses.

During the last few years, developers had proposed various approaches to the conceptual design of Data Warehouses. These approaches try to represent the main multidimensional properties at the conceptual level. However, the conceptual modeling approaches for Data Warehouses should also address other relevant aspects such as initial user requirements, system behavior, available data sources, and specific issues related to automatic generation of the database schemas.

Due to the variation and differences between the proposed approaches for the conceptual design of Data Warehouses, this document aims to present these conceptual modeling approaches, showing the features, advantages and disadvantages of each. To be able to evaluate and compare these conceptual modeling approaches, this document will detail the evaluation criteria upon which different types of conceptual models are evaluated.

Based on the comparison of the conceptual models, we selected the best model up to our knowledge and enhanced this model to support some missing multidimensional modeling properties and also to support building temporal data warehouses. Then, we implemented a prototype CASE tool to support designing conceptual models for data warehouses using our enhanced conceptual model. Finally, we converted the conceptual schema resulted from the prototype CASE tool into XML to facilitate the transformation of the conceptual schema into a DBMS.