A new relaxed memory consistency model for shared-memory multiprocessors with parallel-multithreaded processing elements

Chao-Chin Wu, Cheng Chen

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

The release consistency model is the generally accepted hardware-centric relaxed memory consistency model because of its performance and implementation complexity. By extending the release consistency model, in this paper, we propose a hardware-centric memory consistency model particularly for shared-memory multiprocessor systems with parallel-multithreaded processing elements. The new model uses a new categorization for memory references and utilizes the feature of parallel multithreaded processors (PMPs). We further partition acquire and release references into three sub-categories: one for lock-unlock pairs, one for barrier synchronization, and the last for others. According to the semantic of each synchronization primitive, each sub-category has its own relaxed restrictions. On the other hand, the feature of a PMP is that it is capable of executing more than one thread at the same time, where all parallel threads share only one cache hierarchy. Under the new model, we can use dual write-caches to reduce write traffic and synchronization time. We have used five benchmarks in the SPLASH suite to evaluate the performance gain for the new model. According to the simulation results, the new model is superior to the release consistency model at best by about 11%.

Original languageEnglish
Pages (from-to)785-808
Number of pages24
JournalJournal of Information Science and Engineering
Volume14
Issue number4
Publication statusPublished - 1998 Dec 1

Fingerprint

Data storage equipment
Processing
Synchronization
Computer hardware
hardware
performance
Computer systems
Semantics
semantics
traffic
simulation

All Science Journal Classification (ASJC) codes

  • Software
  • Human-Computer Interaction
  • Hardware and Architecture
  • Library and Information Sciences
  • Computational Theory and Mathematics

Cite this

@article{a47b6766f81d4a70b51b88fb69aa50bc,
title = "A new relaxed memory consistency model for shared-memory multiprocessors with parallel-multithreaded processing elements",
abstract = "The release consistency model is the generally accepted hardware-centric relaxed memory consistency model because of its performance and implementation complexity. By extending the release consistency model, in this paper, we propose a hardware-centric memory consistency model particularly for shared-memory multiprocessor systems with parallel-multithreaded processing elements. The new model uses a new categorization for memory references and utilizes the feature of parallel multithreaded processors (PMPs). We further partition acquire and release references into three sub-categories: one for lock-unlock pairs, one for barrier synchronization, and the last for others. According to the semantic of each synchronization primitive, each sub-category has its own relaxed restrictions. On the other hand, the feature of a PMP is that it is capable of executing more than one thread at the same time, where all parallel threads share only one cache hierarchy. Under the new model, we can use dual write-caches to reduce write traffic and synchronization time. We have used five benchmarks in the SPLASH suite to evaluate the performance gain for the new model. According to the simulation results, the new model is superior to the release consistency model at best by about 11{\%}.",
author = "Chao-Chin Wu and Cheng Chen",
year = "1998",
month = "12",
day = "1",
language = "English",
volume = "14",
pages = "785--808",
journal = "Journal of Information Science and Engineering",
issn = "1016-2364",
publisher = "Institute of Information Science",
number = "4",

}

TY - JOUR

T1 - A new relaxed memory consistency model for shared-memory multiprocessors with parallel-multithreaded processing elements

AU - Wu, Chao-Chin

AU - Chen, Cheng

PY - 1998/12/1

Y1 - 1998/12/1

N2 - The release consistency model is the generally accepted hardware-centric relaxed memory consistency model because of its performance and implementation complexity. By extending the release consistency model, in this paper, we propose a hardware-centric memory consistency model particularly for shared-memory multiprocessor systems with parallel-multithreaded processing elements. The new model uses a new categorization for memory references and utilizes the feature of parallel multithreaded processors (PMPs). We further partition acquire and release references into three sub-categories: one for lock-unlock pairs, one for barrier synchronization, and the last for others. According to the semantic of each synchronization primitive, each sub-category has its own relaxed restrictions. On the other hand, the feature of a PMP is that it is capable of executing more than one thread at the same time, where all parallel threads share only one cache hierarchy. Under the new model, we can use dual write-caches to reduce write traffic and synchronization time. We have used five benchmarks in the SPLASH suite to evaluate the performance gain for the new model. According to the simulation results, the new model is superior to the release consistency model at best by about 11%.

AB - The release consistency model is the generally accepted hardware-centric relaxed memory consistency model because of its performance and implementation complexity. By extending the release consistency model, in this paper, we propose a hardware-centric memory consistency model particularly for shared-memory multiprocessor systems with parallel-multithreaded processing elements. The new model uses a new categorization for memory references and utilizes the feature of parallel multithreaded processors (PMPs). We further partition acquire and release references into three sub-categories: one for lock-unlock pairs, one for barrier synchronization, and the last for others. According to the semantic of each synchronization primitive, each sub-category has its own relaxed restrictions. On the other hand, the feature of a PMP is that it is capable of executing more than one thread at the same time, where all parallel threads share only one cache hierarchy. Under the new model, we can use dual write-caches to reduce write traffic and synchronization time. We have used five benchmarks in the SPLASH suite to evaluate the performance gain for the new model. According to the simulation results, the new model is superior to the release consistency model at best by about 11%.

UR - http://www.scopus.com/inward/record.url?scp=0032268957&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0032268957&partnerID=8YFLogxK

M3 - Article

VL - 14

SP - 785

EP - 808

JO - Journal of Information Science and Engineering

JF - Journal of Information Science and Engineering

SN - 1016-2364

IS - 4

ER -