Abstract

This report documents the implementation of a data warehousing initiative for the purpose of marketing data analysis. Implementation of this project was divided into two phases. The objective of phase one is to produce a concept-proof prototype. Phase two, of which I took major responsibility, is to generate an actual production system. Major tasks I performed in phase two covered many aspects of the data warehousing life cycle: revised and fine-tuned the conceptual, logical and physical data model; performed database redesign and database sizing; built and rebuilt the database to improve performance; improved data extraction, transformation and loading process; performed database and SQL performance tuning; planned and implemented information presentation with off the shell data access tools. The first part of the report reviews the data warehousing literature by examining its evolution, conceptual model, major architectural components and some critical issues involved. In the second part of the report, the implementation of a marketing data warehouse is examined in details. A system overview is provided along with the logical data model. It then describes the mainframe component, UNIX components, presentation/end user component and the interaction among them. The Appendix provides further technical details of the project.