Lightweight contention management for efficient compare-and-swap operations

David Dice; Danny Hendler; Ilya Mirsky

doi:10.1007/978-3-642-40047-6_60

Lightweight contention management for efficient compare-and-swap operations

David Dice, Danny Hendler, Ilya Mirsky

Department of Computer Science

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

9 Scopus citations

Abstract

Many concurrent data-structure implementations use the well-known compare-and-swap (CAS) operation, supported in hardware by most modern multiprocessor architectures, for inter-thread synchronization. A key weakness of the CAS operation is the degradation in its performance in the presence of memory contention. In this work we study the following question: can software-based contention management improve the efficiency of hardware-provided CAS operations? Our performance evaluation establishes that lightweight contention management support can greatly improve performance under medium and high contention levels while typically incurring only small overhead when contention is low.

Original language	English
Title of host publication	Euro-Par 2013 Parallel Processing - 19th International Conference, Proceedings
Pages	595-606
Number of pages	12
DOIs	https://doi.org/10.1007/978-3-642-40047-6_60
State	Published - 3 Sep 2013
Event	19th International Conference on Parallel Processing, Euro-Par 2013 - Aachen, Germany Duration: 26 Aug 2013 → 30 Aug 2013

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	8097 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	19th International Conference on Parallel Processing, Euro-Par 2013
Country/Territory	Germany
City	Aachen
Period	26/08/13 → 30/08/13

Keywords

Compare-and-swap
concurrent algorithms
contention management

ASJC Scopus subject areas

Theoretical Computer Science
Computer Science (all)

Access to Document

10.1007/978-3-642-40047-6_60

Cite this

Dice, D., Hendler, D., & Mirsky, I. (2013). Lightweight contention management for efficient compare-and-swap operations. In Euro-Par 2013 Parallel Processing - 19th International Conference, Proceedings (pp. 595-606). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 8097 LNCS). https://doi.org/10.1007/978-3-642-40047-6_60

@inproceedings{658ccdd51ff34f6c8f884bd657b2917b,

title = "Lightweight contention management for efficient compare-and-swap operations",

abstract = "Many concurrent data-structure implementations use the well-known compare-and-swap (CAS) operation, supported in hardware by most modern multiprocessor architectures, for inter-thread synchronization. A key weakness of the CAS operation is the degradation in its performance in the presence of memory contention. In this work we study the following question: can software-based contention management improve the efficiency of hardware-provided CAS operations? Our performance evaluation establishes that lightweight contention management support can greatly improve performance under medium and high contention levels while typically incurring only small overhead when contention is low.",

keywords = "Compare-and-swap, concurrent algorithms, contention management",

author = "David Dice and Danny Hendler and Ilya Mirsky",

note = "Funding Information: Partially supported by the Israel Science Foundation (grant number 1227/10) and by the Lynne and William Frankel Center for Computing Science at Ben-Gurion University.; 19th International Conference on Parallel Processing, Euro-Par 2013 ; Conference date: 26-08-2013 Through 30-08-2013",

year = "2013",

month = sep,

day = "3",

doi = "10.1007/978-3-642-40047-6_60",

language = "English",

isbn = "9783642400469",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

pages = "595--606",

booktitle = "Euro-Par 2013 Parallel Processing - 19th International Conference, Proceedings",

}

Dice, D, Hendler, D & Mirsky, I 2013, Lightweight contention management for efficient compare-and-swap operations. in Euro-Par 2013 Parallel Processing - 19th International Conference, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 8097 LNCS, pp. 595-606, 19th International Conference on Parallel Processing, Euro-Par 2013, Aachen, Germany, 26/08/13. https://doi.org/10.1007/978-3-642-40047-6_60

Lightweight contention management for efficient compare-and-swap operations. / Dice, David; Hendler, Danny; Mirsky, Ilya.
Euro-Par 2013 Parallel Processing - 19th International Conference, Proceedings. 2013. p. 595-606 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 8097 LNCS).

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

TY - GEN

T1 - Lightweight contention management for efficient compare-and-swap operations

AU - Dice, David

AU - Hendler, Danny

AU - Mirsky, Ilya

N1 - Funding Information: Partially supported by the Israel Science Foundation (grant number 1227/10) and by the Lynne and William Frankel Center for Computing Science at Ben-Gurion University.

PY - 2013/9/3

Y1 - 2013/9/3

N2 - Many concurrent data-structure implementations use the well-known compare-and-swap (CAS) operation, supported in hardware by most modern multiprocessor architectures, for inter-thread synchronization. A key weakness of the CAS operation is the degradation in its performance in the presence of memory contention. In this work we study the following question: can software-based contention management improve the efficiency of hardware-provided CAS operations? Our performance evaluation establishes that lightweight contention management support can greatly improve performance under medium and high contention levels while typically incurring only small overhead when contention is low.

AB - Many concurrent data-structure implementations use the well-known compare-and-swap (CAS) operation, supported in hardware by most modern multiprocessor architectures, for inter-thread synchronization. A key weakness of the CAS operation is the degradation in its performance in the presence of memory contention. In this work we study the following question: can software-based contention management improve the efficiency of hardware-provided CAS operations? Our performance evaluation establishes that lightweight contention management support can greatly improve performance under medium and high contention levels while typically incurring only small overhead when contention is low.

KW - Compare-and-swap

KW - concurrent algorithms

KW - contention management

UR - http://www.scopus.com/inward/record.url?scp=84883201361&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-40047-6_60

DO - 10.1007/978-3-642-40047-6_60

M3 - Conference contribution

AN - SCOPUS:84883201361

SN - 9783642400469

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 595

EP - 606

BT - Euro-Par 2013 Parallel Processing - 19th International Conference, Proceedings

T2 - 19th International Conference on Parallel Processing, Euro-Par 2013

Y2 - 26 August 2013 through 30 August 2013

ER -

Dice D, Hendler D, Mirsky I. Lightweight contention management for efficient compare-and-swap operations. In Euro-Par 2013 Parallel Processing - 19th International Conference, Proceedings. 2013. p. 595-606. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-642-40047-6_60

Lightweight contention management for efficient compare-and-swap operations

Abstract

Publication series

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this