The use of reduction filters in distributed query optimization.
Date of Award
CC BY-NC-ND 4.0
A major issue that affects the performance of a distributed database management system is the optimal processing of a query involving data from several sites. The problem of distributed query processing is to determine a sequence of operations, called an execution strategy, with the minimum cost. This has been shown to be an NP-Hard problem [Hen80, WC96]. Therefore, most proposed algorithms for processing distributed queries are heuristic, and focus on producing efficient (but suboptimal) strategies that minimize some particular cost of the query. Many proposed solutions use joins, semijoins, a combination of joins and semijoins, and dynamic methods. Solutions that use a filter-based approach have also been proposed. However, the limitations of such approaches include the assumption of a perfect hash function, the restriction of the algorithm to specific query types, and the restriction of the algorithm to a specific number of relations and joining attributes. Therefore, we propose a new filter-based algorithm that can process general queries consisting of an arbitrary number of relations and joining attributes. Also, it does not assume the use of a perfect hash function. (Abstract shortened by UMI.) Paper copy at Leddy Library: Theses & Major Papers - Basement, West Bldg. / Call Number: Thesis1998 .O87. Source: Masters Abstracts International, Volume: 39-02, page: 0531. Adviser: Joan Morrissey. Thesis (M.Sc.)--University of Windsor (Canada), 1998.
Osborn, Wendy Kathleen., "The use of reduction filters in distributed query optimization." (1998). Electronic Theses and Dissertations. 4336.