Exploiting storage redundancy to speed Up randomized shared memory simulations

Abstract

This paper presents and analyses efficient implementations of a so-called direct process on distributed memory machines (DMMs) that yields - a simulation of an n-processor PRAM on an n-processor optical crossbar DMM with delay O(log log n), - a simulation of an n-processor PRAM on an n-processor arbitrary DMM with delay O (formula presented) - an implementation of a static dictionary on an n-processor arbitrary DMM with parallel access time of O(log* n). We further prove a lower bound for executing the above process, showing that our implementations are optimal.