Illinois aggressive coma multiprocessor project (I-ACOMA)

Research output: Contribution to conferencePaper

Abstract

While scalable shared-memory multiprocessors with hardware-assisted cache coherence are relatively easy to program, if truly high-performance is desired, they still require substantial programmer effort. For example, data must be allocated close to the processors that will use them and the application must be tuned so that the working set fits in the caches. This is unfortunate because the most important obstacle to widespread use of parallel computing is the hardship of programming parallel machines. The goal of the I-ACOMA project is to explore how to design a highly programmable high-performance multiprocessor. We focus on a flat-coma scalable multiprocessor supported by a parallelizing compiler. The main issues that we are studying are advanced processor organizations, techniques to handle long memory access latencies, and support for important classes of workloads like databases and scientific applications with loops that cannot be compiler-analyzed. The project also involves building a prototype that includes some of the features discussed.

Original languageEnglish (US)
Pages106-111
Number of pages6
StatePublished - Dec 1 1996
EventProceedings of the 1996 6th Symposium on the Frontiers of Massively Parallel Computing, Frontiers'96 - Annapolis, MD, USA
Duration: Oct 27 1996Oct 31 1996

Other

OtherProceedings of the 1996 6th Symposium on the Frontiers of Massively Parallel Computing, Frontiers'96
CityAnnapolis, MD, USA
Period10/27/9610/31/96

Fingerprint

Data storage equipment
Parallel programming
Parallel processing systems
Computer hardware

ASJC Scopus subject areas

  • Hardware and Architecture

Cite this

Torrellas, J., & Padua, D. A. (1996). Illinois aggressive coma multiprocessor project (I-ACOMA). 106-111. Paper presented at Proceedings of the 1996 6th Symposium on the Frontiers of Massively Parallel Computing, Frontiers'96, Annapolis, MD, USA, .

Illinois aggressive coma multiprocessor project (I-ACOMA). / Torrellas, Josep; Padua, David A.

1996. 106-111 Paper presented at Proceedings of the 1996 6th Symposium on the Frontiers of Massively Parallel Computing, Frontiers'96, Annapolis, MD, USA, .

Research output: Contribution to conferencePaper

Torrellas, J & Padua, DA 1996, 'Illinois aggressive coma multiprocessor project (I-ACOMA)' Paper presented at Proceedings of the 1996 6th Symposium on the Frontiers of Massively Parallel Computing, Frontiers'96, Annapolis, MD, USA, 10/27/96 - 10/31/96, pp. 106-111.
Torrellas J, Padua DA. Illinois aggressive coma multiprocessor project (I-ACOMA). 1996. Paper presented at Proceedings of the 1996 6th Symposium on the Frontiers of Massively Parallel Computing, Frontiers'96, Annapolis, MD, USA, .
Torrellas, Josep ; Padua, David A. / Illinois aggressive coma multiprocessor project (I-ACOMA). Paper presented at Proceedings of the 1996 6th Symposium on the Frontiers of Massively Parallel Computing, Frontiers'96, Annapolis, MD, USA, .6 p.
@conference{73fdee29d9a44da09eb88f91528278ac,
title = "Illinois aggressive coma multiprocessor project (I-ACOMA)",
abstract = "While scalable shared-memory multiprocessors with hardware-assisted cache coherence are relatively easy to program, if truly high-performance is desired, they still require substantial programmer effort. For example, data must be allocated close to the processors that will use them and the application must be tuned so that the working set fits in the caches. This is unfortunate because the most important obstacle to widespread use of parallel computing is the hardship of programming parallel machines. The goal of the I-ACOMA project is to explore how to design a highly programmable high-performance multiprocessor. We focus on a flat-coma scalable multiprocessor supported by a parallelizing compiler. The main issues that we are studying are advanced processor organizations, techniques to handle long memory access latencies, and support for important classes of workloads like databases and scientific applications with loops that cannot be compiler-analyzed. The project also involves building a prototype that includes some of the features discussed.",
author = "Josep Torrellas and Padua, {David A}",
year = "1996",
month = "12",
day = "1",
language = "English (US)",
pages = "106--111",
note = "Proceedings of the 1996 6th Symposium on the Frontiers of Massively Parallel Computing, Frontiers'96 ; Conference date: 27-10-1996 Through 31-10-1996",

}

TY - CONF

T1 - Illinois aggressive coma multiprocessor project (I-ACOMA)

AU - Torrellas, Josep

AU - Padua, David A

PY - 1996/12/1

Y1 - 1996/12/1

N2 - While scalable shared-memory multiprocessors with hardware-assisted cache coherence are relatively easy to program, if truly high-performance is desired, they still require substantial programmer effort. For example, data must be allocated close to the processors that will use them and the application must be tuned so that the working set fits in the caches. This is unfortunate because the most important obstacle to widespread use of parallel computing is the hardship of programming parallel machines. The goal of the I-ACOMA project is to explore how to design a highly programmable high-performance multiprocessor. We focus on a flat-coma scalable multiprocessor supported by a parallelizing compiler. The main issues that we are studying are advanced processor organizations, techniques to handle long memory access latencies, and support for important classes of workloads like databases and scientific applications with loops that cannot be compiler-analyzed. The project also involves building a prototype that includes some of the features discussed.

AB - While scalable shared-memory multiprocessors with hardware-assisted cache coherence are relatively easy to program, if truly high-performance is desired, they still require substantial programmer effort. For example, data must be allocated close to the processors that will use them and the application must be tuned so that the working set fits in the caches. This is unfortunate because the most important obstacle to widespread use of parallel computing is the hardship of programming parallel machines. The goal of the I-ACOMA project is to explore how to design a highly programmable high-performance multiprocessor. We focus on a flat-coma scalable multiprocessor supported by a parallelizing compiler. The main issues that we are studying are advanced processor organizations, techniques to handle long memory access latencies, and support for important classes of workloads like databases and scientific applications with loops that cannot be compiler-analyzed. The project also involves building a prototype that includes some of the features discussed.

UR - http://www.scopus.com/inward/record.url?scp=0030419251&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0030419251&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:0030419251

SP - 106

EP - 111

ER -