The process of metadata modelling in industrial data warehouse environments

Abstract

Modern application landscapes and especially huge enterprise applications, like data warehouses, used for decision support or other analyzing purposes get more and more complex. To manage, use and maintain these systems the need for metadata management has increased. In consequence of new tasks being identified by new groups of data warehouse users, the role of metadata management implies more than simply surf data schemas. It becomes necessary that metadata systems integrate different kinds of metadata and offer different views on the metadata as well. In this paper we discuss the process of identifying metadata model requirements, defining a new metadata model and finally implementing it in a metadata schema. The process is illustrated by a possible metadata model and schema, which were developed to meet the requirements of a complex data warehouse environment in Helsana Versicherungen AG, the largest Swiss insurance company. The paper describes the implementation of the metadata model based on the metadata standards Resource Description Framework (RDF) and RDF Schema (RDFS). The presented model and schema are just one possible solution and are not leading to a universal metadata model. The goal of this paper is to discuss the process of metadata modeling and to help metadata architects to develop their own metadata models and schemas.