The use of multiple Bloom-filters to minimize data transfer during distributed query processing.

Query processing in distributed database system requires the transmission of data between computers in network. Therefore, query optimization in distributed database system is an important research issue. Since employing the optimization for general query is NP-Hard, heuristics are applied to find a cost-effective and efficient processing strategy. The challenge is how to efficiently minimize either transmission time or local processing cost in a query process. In the thesis, we propose a new reduction approach to significantly minimize data transmission time. The algorithm [32] is used to process general queries by simply substituting single Bloom filter that is based on perfect hashing in the reduction approach [32] with multiple Bloom filters which are based on non-perfect hash functions. Our approach aims to minimize data transmission time. The evaluation of our application is against the reducer [32]. An analysis of how the number of Bloom-filters affects the performance of the algorithm [32] is provided in the thesis. An amount of experimental results will be used to evaluate the performance of our reduction approach. Compared to the approach in paper [32], our reduction approach provides a more practical, cost effective and efficient processing query solution. Paper copy at Leddy Library: Theses & Major Papers - Basement, West Bldg. / Call Number: Thesis2000 .W36. Source: Masters Abstracts International, Volume: 40-03, page: 0729. Thesis (M.Sc.)--University of Windsor (Canada), 2001