This will be a presentation on my research field to approach the ways of Materialized views evolution in Data Warehouse. In addition to this, an introduction to background of Data Warehouse and the research issues in Data Warehouse will be given briefly.

There is a great deal of information in the Data Warehouse. Decision making based on the Data Warehouse, such as on-line analytical processing (OLAP), puts some different and complex requirements on database technology compared to traditional on-line transaction processing applications. Warehouse applications therefore build a large number of Materialized views to assist the Warehouse to perform well. However, the various Materialized views related to each other constitute a complicated architecture and, moreover, Data Warehouse is a dynamic environment, when user query requirement changes, the< Materialized views corresponding to the new query set will change. How should we adjust the Materialized views with respect to the new queries when a new query is added? In this paper, we first tackle this problem. An algorithm is proposed to deal with this kind of problem when adding a new query into the query set. The aim of the algorithm is to determine how to efficiently get the new Materialized views by making use of the existing Materialized views without recomputing the whole nodes.