DiSTM:: A software transactional memory framework for clusters

Christos Kotselidis; Mohammad Ansari; Kim Jarvis; Mikel Luján; Chris Kirkham; Ian Watson

doi:10.1109/ICPP.2008.59

DiSTM: A software transactional memory framework for clusters

Christos Kotselidis, Mohammad Ansari, Kim Jarvis, Mikel Luján, Chris Kirkham, Ian Watson

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

While Transactional Memory (TM) research on shared-memory chip multiprocessors has been flourishing over the last years, limited research has been conducted in the cluster domain. In this paper, we introduce a research platform for exploiting software TM on clusters. The Distributed Software Transactional Memory (DiSTM) system has been designed for easy prototyping of TM coherence protocols and it does not rely on a software or hardware implementation of distributed shared memory. Three TM coherence protocols have been implemented and evaluated with established TM benchmarks. The decentralized Transactional Coherence and Consistency protocol has been compared against two centralized protocols that utilize leases. Results indicate that depending on network congestion and amount of contention different protocols perform better. © 2008 IEEE.

Original language	English
Title of host publication	Proceedings of the International Conference on Parallel Processing\|Proc. Int. Conf. Parallel Process.
Publisher	IEEE Computer Society
Pages	51-58
Number of pages	7
ISBN (Print)	9780769533742
DOIs	https://doi.org/10.1109/ICPP.2008.59
Publication status	Published - 2008
Event	37th International Conference on Parallel Processing, ICPP 2008 - Portland, OR Duration: 1 Jul 2008 → … http://dblp.uni-trier.de/db/conf/icpp/icpp2008.html#KotselidisAJLKW08http://dblp.uni-trier.de/rec/bibtex/conf/icpp/KotselidisAJLKW08.xmlhttp://dblp.uni-trier.de/rec/bibtex/conf/icpp/KotselidisAJLKW08

Conference

Conference	37th International Conference on Parallel Processing, ICPP 2008
City	Portland, OR
Period	1/07/08 → …
Internet address	http://dblp.uni-trier.de/db/conf/icpp/icpp2008.html#KotselidisAJLKW08http://dblp.uni-trier.de/rec/bibtex/conf/icpp/KotselidisAJLKW08.xmlhttp://dblp.uni-trier.de/rec/bibtex/conf/icpp/KotselidisAJLKW08

Access to Document

10.1109/ICPP.2008.59

Cite this

@inproceedings{2e8a9be4fc87423f8ad00cb768752b99,

title = "DiSTM:: A software transactional memory framework for clusters",

abstract = "While Transactional Memory (TM) research on shared-memory chip multiprocessors has been flourishing over the last years, limited research has been conducted in the cluster domain. In this paper, we introduce a research platform for exploiting software TM on clusters. The Distributed Software Transactional Memory (DiSTM) system has been designed for easy prototyping of TM coherence protocols and it does not rely on a software or hardware implementation of distributed shared memory. Three TM coherence protocols have been implemented and evaluated with established TM benchmarks. The decentralized Transactional Coherence and Consistency protocol has been compared against two centralized protocols that utilize leases. Results indicate that depending on network congestion and amount of contention different protocols perform better. {\textcopyright} 2008 IEEE.",

author = "Christos Kotselidis and Mohammad Ansari and Kim Jarvis and Mikel Luj{\'a}n and Chris Kirkham and Ian Watson",

year = "2008",

doi = "10.1109/ICPP.2008.59",

language = "English",

isbn = "9780769533742",

pages = "51--58",

booktitle = "Proceedings of the International Conference on Parallel Processing|Proc. Int. Conf. Parallel Process.",

publisher = "IEEE Computer Society ",

address = "United States",

note = "37th International Conference on Parallel Processing, ICPP 2008 ; Conference date: 01-07-2008",

url = "http://dblp.uni-trier.de/db/conf/icpp/icpp2008.html#KotselidisAJLKW08http://dblp.uni-trier.de/rec/bibtex/conf/icpp/KotselidisAJLKW08.xmlhttp://dblp.uni-trier.de/rec/bibtex/conf/icpp/KotselidisAJLKW08",

}

Kotselidis, C, Ansari, M, Jarvis, K, Luján, M, Kirkham, C & Watson, I 2008, DiSTM: A software transactional memory framework for clusters. in Proceedings of the International Conference on Parallel Processing|Proc. Int. Conf. Parallel Process.. IEEE Computer Society , pp. 51-58, 37th International Conference on Parallel Processing, ICPP 2008, Portland, OR, 1/07/08. https://doi.org/10.1109/ICPP.2008.59

TY - GEN

T1 - DiSTM:

T2 - 37th International Conference on Parallel Processing, ICPP 2008

AU - Kotselidis, Christos

AU - Ansari, Mohammad

AU - Jarvis, Kim

AU - Luján, Mikel

AU - Kirkham, Chris

AU - Watson, Ian

PY - 2008

Y1 - 2008

N2 - While Transactional Memory (TM) research on shared-memory chip multiprocessors has been flourishing over the last years, limited research has been conducted in the cluster domain. In this paper, we introduce a research platform for exploiting software TM on clusters. The Distributed Software Transactional Memory (DiSTM) system has been designed for easy prototyping of TM coherence protocols and it does not rely on a software or hardware implementation of distributed shared memory. Three TM coherence protocols have been implemented and evaluated with established TM benchmarks. The decentralized Transactional Coherence and Consistency protocol has been compared against two centralized protocols that utilize leases. Results indicate that depending on network congestion and amount of contention different protocols perform better. © 2008 IEEE.

AB - While Transactional Memory (TM) research on shared-memory chip multiprocessors has been flourishing over the last years, limited research has been conducted in the cluster domain. In this paper, we introduce a research platform for exploiting software TM on clusters. The Distributed Software Transactional Memory (DiSTM) system has been designed for easy prototyping of TM coherence protocols and it does not rely on a software or hardware implementation of distributed shared memory. Three TM coherence protocols have been implemented and evaluated with established TM benchmarks. The decentralized Transactional Coherence and Consistency protocol has been compared against two centralized protocols that utilize leases. Results indicate that depending on network congestion and amount of contention different protocols perform better. © 2008 IEEE.

U2 - 10.1109/ICPP.2008.59

DO - 10.1109/ICPP.2008.59

M3 - Conference contribution

SN - 9780769533742

SP - 51

EP - 58

BT - Proceedings of the International Conference on Parallel Processing|Proc. Int. Conf. Parallel Process.

PB - IEEE Computer Society

Y2 - 1 July 2008

ER -

DiSTM: A software transactional memory framework for clusters

Abstract

Conference

Access to Document

Fingerprint

Cite this