login

Optimally universal parallel computers

Philosophical Transactions of the Royal Society of London Series A Mathematical and Physical SciencesPublished 26 September 1988
Leslie G. Valiant
Citations12

TL;DR

It is shown that any program written for the idealized shared-memory model of parallel computation can be simulated on a hypercube architecture with only constant factor inefficiency, provided that the original program has a certain amount of parallel slackness.

Abstract

Abstract It is shown that any program written for the idealized shared-memory model of parallel computation can be simulated on a hypercube architecture with only constant factor inefficiency, provided that the original program has a certain amount of parallel slackness.

Keywords

Computer Science