Self-maintainable Data Warehouse Views Using Differential Files

Data warehouses materialize views in order to provide fast access to information, but the correctness also should be maintained. The notion of selfmaintenance can be a goal that defined as maintaining views by materializing supplementary data. Incremental update mechanisms can now be prospective way to the goal. A discussion with extensive literature is generated on several issues with maintaining data warehouse views. There are issues related to the data warehouse view maintenance: algebraic approaches, the concurrency control problem, database rule related problems. Motivating example derived from TPC-D database is suggested in terms of incremental base relation method, auxiliary view method, and differential method respectively. After quantitative dealings and qualitative measures are reviewed, it is concluded that the differential method is superior to the others in that it is a true selfmaintenance method and it does not burden the current base relations.This work was supported by the Korea Science and Engineering Foundation (KOSEF) through the Advanced Information Technology Research Center (AITrc), and the first author was supported by postdoctoral fellowship program from KOSEF