A global heuristic for distributed join operations

This paper deals with the problem of query optimization in relational distributed data-base systems. It concentrates on the development of heuristic procedures that attempt to minimize the communication costs incurred by the distributed processing of queries. In particular, the paper deals with a class of heuristics that use a semi-join strategy as the mechanism for communication cost reduction. These heuristics are classified into two types-local and global heuristics. The global heuristic proposed in this paper is based on an optimal solution to a mathematical model of a simplified version of the problem. It is shown that the global heuristic can generate beneficial semi-join operations not included in local heuristic solutions.