The NYU Ultracomputer - Designing a mimd, shared-memory parallel machine (extended abstract)

Allan Gottlieb, Ralph Grishman, Clyde P. Kruskal, Kevin P. McAuliffe, Larry Rudolph, Marc Snir

Research output: Contribution to journalConference article

Abstract

We present the design for the NYU Ultracomputer, a shared-memory MIMD parallel machine composed of thousands of autonomous processing elements. This machine uses an enhanced message switching network with the geometry of an Omega-network to approximate the ideal behavior of Schwartz's paracomputer model of computation and to implement efficiently the important fetch-and-add synchronization primitive. We outline the hardware that would be required to build a 4096 processor system using 1990's technology. We also discuss system software issues, and present analytic studies of the network performance. Finally, we include a sample of our effort to implement and simulate parallel variants of important scientific programs.

Original languageEnglish (US)
Pages (from-to)27-42
Number of pages16
JournalProceedings - International Symposium on Computer Architecture
StatePublished - Apr 26 1982
Event9th Annual Symposium on Computer Architecture, ISCA 1982 - Austin, United States
Duration: Apr 26 1982Apr 29 1982

Fingerprint

Switching networks
Network performance
Synchronization
Hardware
Data storage equipment
Geometry
Processing

ASJC Scopus subject areas

  • Hardware and Architecture

Cite this

The NYU Ultracomputer - Designing a mimd, shared-memory parallel machine (extended abstract). / Gottlieb, Allan; Grishman, Ralph; Kruskal, Clyde P.; McAuliffe, Kevin P.; Rudolph, Larry; Snir, Marc.

In: Proceedings - International Symposium on Computer Architecture, 26.04.1982, p. 27-42.

Research output: Contribution to journalConference article

@article{3fbfc1d4c531417cac3e963b07b0d7e7,
title = "The NYU Ultracomputer - Designing a mimd, shared-memory parallel machine (extended abstract)",
abstract = "We present the design for the NYU Ultracomputer, a shared-memory MIMD parallel machine composed of thousands of autonomous processing elements. This machine uses an enhanced message switching network with the geometry of an Omega-network to approximate the ideal behavior of Schwartz's paracomputer model of computation and to implement efficiently the important fetch-and-add synchronization primitive. We outline the hardware that would be required to build a 4096 processor system using 1990's technology. We also discuss system software issues, and present analytic studies of the network performance. Finally, we include a sample of our effort to implement and simulate parallel variants of important scientific programs.",
author = "Allan Gottlieb and Ralph Grishman and Kruskal, {Clyde P.} and McAuliffe, {Kevin P.} and Larry Rudolph and Marc Snir",
year = "1982",
month = "4",
day = "26",
language = "English (US)",
pages = "27--42",
journal = "Conference Proceedings - Annual International Symposium on Computer Architecture, ISCA",
issn = "1063-6897",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - JOUR

T1 - The NYU Ultracomputer - Designing a mimd, shared-memory parallel machine (extended abstract)

AU - Gottlieb, Allan

AU - Grishman, Ralph

AU - Kruskal, Clyde P.

AU - McAuliffe, Kevin P.

AU - Rudolph, Larry

AU - Snir, Marc

PY - 1982/4/26

Y1 - 1982/4/26

N2 - We present the design for the NYU Ultracomputer, a shared-memory MIMD parallel machine composed of thousands of autonomous processing elements. This machine uses an enhanced message switching network with the geometry of an Omega-network to approximate the ideal behavior of Schwartz's paracomputer model of computation and to implement efficiently the important fetch-and-add synchronization primitive. We outline the hardware that would be required to build a 4096 processor system using 1990's technology. We also discuss system software issues, and present analytic studies of the network performance. Finally, we include a sample of our effort to implement and simulate parallel variants of important scientific programs.

AB - We present the design for the NYU Ultracomputer, a shared-memory MIMD parallel machine composed of thousands of autonomous processing elements. This machine uses an enhanced message switching network with the geometry of an Omega-network to approximate the ideal behavior of Schwartz's paracomputer model of computation and to implement efficiently the important fetch-and-add synchronization primitive. We outline the hardware that would be required to build a 4096 processor system using 1990's technology. We also discuss system software issues, and present analytic studies of the network performance. Finally, we include a sample of our effort to implement and simulate parallel variants of important scientific programs.

UR - http://www.scopus.com/inward/record.url?scp=85057191630&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85057191630&partnerID=8YFLogxK

M3 - Conference article

AN - SCOPUS:85057191630

SP - 27

EP - 42

JO - Conference Proceedings - Annual International Symposium on Computer Architecture, ISCA

JF - Conference Proceedings - Annual International Symposium on Computer Architecture, ISCA

SN - 1063-6897

ER -