Distance: a New Metric for Controlling Granularity for Parallel Execution
Kish Shen, Vítor Santos Costa and Andy King
University of Manchester / DCC-FC & LIACC, Universidade do Porto / University of Kent at Canterbury
We argue in this paper that the estimation of complexity on its own is not an ideal metric for improving the performance of parallel programs through granularity control. We present a new metric for measuring granularity, based on a notion of distance. We present some initial results with two very simple methods of using this metric for granularity control. We then discuss how more sophisticated granularity control methods can be devised using the new metric.
Keywords: Parallel Logic Programming, And-Or Parallelism, Program Analysis.