Cashmere-VLM: Remote memory paging for software distributed shared memory

Sandhya Dwarkadas*, Nikolaos Hardavellas, Leonidas Kontothanassis, Rishiyur Nikhil, Robert Stets

*Corresponding author for this work

Research output: Contribution to journalConference article

23 Citations (Scopus)

Abstract

Software distributed shared memory (DSM) systems have successfully provided the illusion of shared memory on distributed memory machines. However, most software DSM systems use the main memory of each machine as a level in a cache hierarchy, replicating copies of shared data in local memory. Since computer memories tend to be much larger than caches, DSM systems have largely ignored memory capacity issues, assuming there is always enough space in main memory in which to replicate data. Applications that access data that exceeds the capacity available in local memory will page to disk, resulting in reduced performance. We have developed a software DSM system based on Cashmere that takes advantage of system-wide memory resources in order to reduce or eliminate paging overhead. Experimental results on a 4-node, 16-processor AlphaServer system demonstrate the improvement in performance using the enhanced software DSM system for applications with large data sets.

Original languageEnglish (US)
Pages (from-to)153-159
Number of pages7
JournalProceedings of the International Parallel Processing Symposium, IPPS
StatePublished - Jan 1 1999
EventProceedings of the 1999 13th International Parallel Processing Symposium and 10th Symposium on Parallel and Distributed Processing - San Juan
Duration: Apr 12 1999Apr 16 1999

Fingerprint

Data storage equipment
Computer systems

ASJC Scopus subject areas

  • Hardware and Architecture

Cite this

Dwarkadas, Sandhya ; Hardavellas, Nikolaos ; Kontothanassis, Leonidas ; Nikhil, Rishiyur ; Stets, Robert. / Cashmere-VLM : Remote memory paging for software distributed shared memory. In: Proceedings of the International Parallel Processing Symposium, IPPS. 1999 ; pp. 153-159.
@article{984662d97ff047d68e4ffa24ba98bcce,
title = "Cashmere-VLM: Remote memory paging for software distributed shared memory",
abstract = "Software distributed shared memory (DSM) systems have successfully provided the illusion of shared memory on distributed memory machines. However, most software DSM systems use the main memory of each machine as a level in a cache hierarchy, replicating copies of shared data in local memory. Since computer memories tend to be much larger than caches, DSM systems have largely ignored memory capacity issues, assuming there is always enough space in main memory in which to replicate data. Applications that access data that exceeds the capacity available in local memory will page to disk, resulting in reduced performance. We have developed a software DSM system based on Cashmere that takes advantage of system-wide memory resources in order to reduce or eliminate paging overhead. Experimental results on a 4-node, 16-processor AlphaServer system demonstrate the improvement in performance using the enhanced software DSM system for applications with large data sets.",
author = "Sandhya Dwarkadas and Nikolaos Hardavellas and Leonidas Kontothanassis and Rishiyur Nikhil and Robert Stets",
year = "1999",
month = "1",
day = "1",
language = "English (US)",
pages = "153--159",
journal = "Proceedings of the International Parallel Processing Symposium, IPPS",
issn = "1063-7133",

}

Cashmere-VLM : Remote memory paging for software distributed shared memory. / Dwarkadas, Sandhya; Hardavellas, Nikolaos; Kontothanassis, Leonidas; Nikhil, Rishiyur; Stets, Robert.

In: Proceedings of the International Parallel Processing Symposium, IPPS, 01.01.1999, p. 153-159.

Research output: Contribution to journalConference article

TY - JOUR

T1 - Cashmere-VLM

T2 - Remote memory paging for software distributed shared memory

AU - Dwarkadas, Sandhya

AU - Hardavellas, Nikolaos

AU - Kontothanassis, Leonidas

AU - Nikhil, Rishiyur

AU - Stets, Robert

PY - 1999/1/1

Y1 - 1999/1/1

N2 - Software distributed shared memory (DSM) systems have successfully provided the illusion of shared memory on distributed memory machines. However, most software DSM systems use the main memory of each machine as a level in a cache hierarchy, replicating copies of shared data in local memory. Since computer memories tend to be much larger than caches, DSM systems have largely ignored memory capacity issues, assuming there is always enough space in main memory in which to replicate data. Applications that access data that exceeds the capacity available in local memory will page to disk, resulting in reduced performance. We have developed a software DSM system based on Cashmere that takes advantage of system-wide memory resources in order to reduce or eliminate paging overhead. Experimental results on a 4-node, 16-processor AlphaServer system demonstrate the improvement in performance using the enhanced software DSM system for applications with large data sets.

AB - Software distributed shared memory (DSM) systems have successfully provided the illusion of shared memory on distributed memory machines. However, most software DSM systems use the main memory of each machine as a level in a cache hierarchy, replicating copies of shared data in local memory. Since computer memories tend to be much larger than caches, DSM systems have largely ignored memory capacity issues, assuming there is always enough space in main memory in which to replicate data. Applications that access data that exceeds the capacity available in local memory will page to disk, resulting in reduced performance. We have developed a software DSM system based on Cashmere that takes advantage of system-wide memory resources in order to reduce or eliminate paging overhead. Experimental results on a 4-node, 16-processor AlphaServer system demonstrate the improvement in performance using the enhanced software DSM system for applications with large data sets.

UR - http://www.scopus.com/inward/record.url?scp=0032633386&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0032633386&partnerID=8YFLogxK

M3 - Conference article

AN - SCOPUS:0032633386

SP - 153

EP - 159

JO - Proceedings of the International Parallel Processing Symposium, IPPS

JF - Proceedings of the International Parallel Processing Symposium, IPPS

SN - 1063-7133

ER -