Accession Number : ADA327912

Title :   How to Share Memory in a Distributed System.

Corporate Author : STANFORD UNIV CA DEPT OF COMPUTER SCIENCE

Personal Author(s) : Upfal, Eli ; Wigderson, Avi

PDF Url : ADA327912

Report Date : OCT 1984

Pagination or Media Count : 21

Abstract : We study the power of shared memory in models of parallel computation. We describe a novel distributed data structure that eliminates the need for shared memory without significantly increasing the run time of the parallel computation. More specifically we show how a complete network of processors can deterministicly simulate one PRAM step in O(log n(loglog n)2) time, when both models use n processors, and the size of the PRAM's shared memory is polynomial in n. The best previously known upper bound was the trivial O(n). We also establish that this upper bounds is nearly optimal. We prove that an online simulation of T PRAM steps by a complete network of processors requires omega(T log n) time/loglog n. A simple consequence of the upper bound is that an Ultracomputer (the only currently feasible general purpose parallel machine), can simulate one step of a PRAM (the most convenient parallel model to program), in ()((log n loglog n)2) steps.

Descriptors :   *DISTRIBUTED DATA PROCESSING, *PARALLEL PROCESSING, *TIME SHARING, DATA BASES, SIMULATION, COMPUTATIONS, MODELS, NETWORKS, PROCESSING EQUIPMENT, MEMORY DEVICES, POLYNOMIALS, POWER, PARALLEL ORIENTATION, ONLINE SYSTEMS.

Subject Categories : Computer Systems

Distribution Statement : APPROVED FOR PUBLIC RELEASE