A Holistic Approach for Managing Requirements of Data Warehouse Systems
Abstract:
Quantitive evaluations show that many data warehouse projects fail at an
alarming rate, wasting all the time, money, and effort spent on them. Cost
overruns and schedule delays are attributed in large part, to problems
associated with requirements. Therefore, a crucial success factor for the
implementation of data warehouse systems is efficient requirements management.
In this paper we introduce the easyREMOTEDWH (easy Requirements Modeling
Technique for Data WareHouses) requirements management approach, which
facilitates an incremental and iterative development of complete, correct,
feasible, necessary, prioritized, unambiguous, and verifiable data warehouse
requirements. As a foundation we present a requirements specification
template with consideration of different stakeholder perspectives and the
typical characteristics of data warehouse systems. We discuss traditional
requirements management activities (change control, version control, status
tracking, requirements tracing) in the context of data warehouse environments
and extend these traditional activities to support more efficient requirements
documentation and visibility.
Authors:
Josef Schiefer
IBM Watson Research Center, New York, USA.
Beate List
Institute of Software Technology, Vienna University of Technology, Austria.
Robert M. Bruckner
Institute of Software Technology, Vienna University of Technology, Austria.
Publishing Information:
In Proceedings of the 8th Americas Conference on Information Systems (AMCIS 2002),
pp. 77-87, Dallas, Texas, USA, August 2002.