No abstract

No abstract

No abstract

Multidimensional database technology is becoming more and more important in conjunction with data warehouses and OLAP analysis. What is still lacking is a commonly accepted formal foundation . Such a model can serve as a basis for future research and standardization. Recently a multitude of interesting proposals on this topic have been published. OLAP applications have some special requirements that do not apply to other areas of multidimensional analysis (e.g. GIS, PACS). In this paper we list requirements that a formal model and a corresponding query language must fulfill to be suitable for OLAP. We compare four approaches that come closest to our requirements. After a brief description we discuss their suitability as a formal foundation for OLAP, thus providing a systematic overview. Finally, we propose directions for further research.

Generating tool specific schemata and configuration information for OLAP database tools from conceptual graphical models is an important prerequisite for a comprehensive tool support for computer aided data warehouse engineering (CAWE). This paper describes the design and implementation of such a generation component in the context of our BabelFish data warehouse design tool environment. It identifies the principal issues that are involved in the design and implementation of such a component and discusses possible solutions. The paper lists typical mismatches between the data model of commercial OLAP tools and conceptual graphical modeling notations, and proposes methods to overcome these expressive differences during the generation process. Further topics are the use of graph grammars for specifying and parsing graphical MD schema descriptions and the integration of the generation process into a metadata centered modeling tool environment.