Metadata management for data warehousing: between vision and reality

Capturing, representing and processing metadata promises to facilitate the management, consistent use and understanding of data and thus better support the exploitation of masses of information that is available online today. Despite the increasing interest in metadata management, its purpose, requirements and problems are still not clear. This is particularly true in the area of data warehousing. The reasons are multiple. Compared to the past, today's metadata management considers a significantly larger spectrum of information (including even certain pieces of programs). Moreover, metadata are produced by various tools and reside in different sources which need to be integrated in order to ensure consistency and provide uniform access, impact analysis and data tracking. Existing work has only partially covered some of these aspects. The paper summarizes the most important issues of metadata management for data warehousing, including the role of metadata and solved and unsolved problems of the available solutions. The design of an appropriate information model, metadata integration and advanced user interaction facilities are crucial questions to be answered.