© 1988 by British Computer Society
| ||||||||||||||||||||||||||||||||||||||||||||||||||||
A Query Processing Algorithm for Distributed Relational Database Systems

1 Department of Computer Science, Virginia Polytechnic Institute and State University, 2990 Telestar Court, Falls Church, Virginia 22042, USA, 2 Department of Industrial Engineering and Operations Research, Virginia Polytechnic Institute and State University, 2990 Telestar Court, Falls Church, Virginia 22042, USA
A heuristic-based algorithm is presented which determines the combination of relational algebra operators and their corresponding processing sites in the network, in response to a query or class of queries. A technique is proposed for the computation of the resulting relation size after a projection, a selection and a join on non-key attributes or the estimation of the resulting relation size after a join on key attributes. This technique supplies the algorithm with the necessary input for determining which relations are to be sent across the network for processing. First, we describe how this algorithm differs from what is reported, in the literature. Second, the algorithm is formulated and, third its implementation is described through an example. Finally, we summarise our findings and discuss future work.
Received August 1985. revised November 1986.
* Department of Computer Science, Virginia Polytechnic Institute and State University, 2990 Telestar Court, Falls Church, Virginia 22042, USA
Department of Industrial Engineering and Operations Research, Virginia Polytechnic Institute and State University, 2990 Telestar Court, Falls Church, Virginia 22042, USA