Work dealing

Danny Hendler; Nir Shavit

doi:10.1145/564870.564900

Work dealing

Danny Hendler, Nir Shavit

Research output: Contribution to conference › Paper › peer-review

12 Scopus citations

Abstract

This paper introduces work-dealing, a new algorithm for "locality oriented" load distribution on small scale shared memory multi-processors. Its key feature is an unprecedented low overhead mechanism (only a couple of loads and stores per operation, and no costly compare-and-swaps) for dealing-out work to processors in a globally balanced way. We believe that for applications in which work-items have process affinity, especially applications running in dedicated mode ("stand alone"), work-dealing could prove a worthy alternative to the popular work-stealing paradigm.

Original language	English
Pages	164-172
Number of pages	9
DOIs	https://doi.org/10.1145/564870.564900
State	Published - 1 Jan 2002
Externally published	Yes
Event	Fourteenth Annual ACM Symposium on Parallel Algorithms and Architectures - Winnipeg, MAN., Canada Duration: 10 Aug 2002 → 13 Aug 2002

Conference

Conference	Fourteenth Annual ACM Symposium on Parallel Algorithms and Architectures
Country/Territory	Canada
City	Winnipeg, MAN.
Period	10/08/02 → 13/08/02

Keywords

Distributed
Load-balancing
Wait-free

ASJC Scopus subject areas

Software
Safety, Risk, Reliability and Quality

Access to Document

10.1145/564870.564900

Cite this

@conference{ff392c0abf8740548d11526512ed9607,

title = "Work dealing",

abstract = "This paper introduces work-dealing, a new algorithm for {"}locality oriented{"} load distribution on small scale shared memory multi-processors. Its key feature is an unprecedented low overhead mechanism (only a couple of loads and stores per operation, and no costly compare-and-swaps) for dealing-out work to processors in a globally balanced way. We believe that for applications in which work-items have process affinity, especially applications running in dedicated mode ({"}stand alone{"}), work-dealing could prove a worthy alternative to the popular work-stealing paradigm.",

keywords = "Distributed, Load-balancing, Wait-free",

author = "Danny Hendler and Nir Shavit",

year = "2002",

month = jan,

day = "1",

doi = "10.1145/564870.564900",

language = "English",

pages = "164--172",

note = "Fourteenth Annual ACM Symposium on Parallel Algorithms and Architectures ; Conference date: 10-08-2002 Through 13-08-2002",

}

TY - CONF

T1 - Work dealing

AU - Hendler, Danny

AU - Shavit, Nir

PY - 2002/1/1

Y1 - 2002/1/1

N2 - This paper introduces work-dealing, a new algorithm for "locality oriented" load distribution on small scale shared memory multi-processors. Its key feature is an unprecedented low overhead mechanism (only a couple of loads and stores per operation, and no costly compare-and-swaps) for dealing-out work to processors in a globally balanced way. We believe that for applications in which work-items have process affinity, especially applications running in dedicated mode ("stand alone"), work-dealing could prove a worthy alternative to the popular work-stealing paradigm.

AB - This paper introduces work-dealing, a new algorithm for "locality oriented" load distribution on small scale shared memory multi-processors. Its key feature is an unprecedented low overhead mechanism (only a couple of loads and stores per operation, and no costly compare-and-swaps) for dealing-out work to processors in a globally balanced way. We believe that for applications in which work-items have process affinity, especially applications running in dedicated mode ("stand alone"), work-dealing could prove a worthy alternative to the popular work-stealing paradigm.

KW - Distributed

KW - Load-balancing

KW - Wait-free

UR - http://www.scopus.com/inward/record.url?scp=0036954486&partnerID=8YFLogxK

U2 - 10.1145/564870.564900

DO - 10.1145/564870.564900

M3 - Paper

AN - SCOPUS:0036954486

SP - 164

EP - 172

T2 - Fourteenth Annual ACM Symposium on Parallel Algorithms and Architectures

Y2 - 10 August 2002 through 13 August 2002

ER -

Work dealing

Abstract

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this