Parallel searching for binary Cartesian product files
Yuan Y. Sung · 1985
The problem of distributing buckets in a file among m disks to facilitate parallel searching for a set of queries is analysed in this paper. We are particularly concerned with the file distribution problem for binary Cartesian product files, and partial match queries. A method is proposed and shown to be strict optimal under certain conditions. The performance of the proposed method is compared with those of an “ideal” strict optimal and Du's heuristic allocation method.