Work dealing

Danny Hendler, Nir Shavit

Research output: Contribution to conferencePaperpeer-review

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 languageEnglish
Pages164-172
Number of pages9
DOIs
StatePublished - 2002
EventFourteenth Annual ACM Symposium on Parallel Algorithms and Architectures - Winnipeg, MAN., Canada
Duration: 10 Aug 200213 Aug 2002

Conference

ConferenceFourteenth Annual ACM Symposium on Parallel Algorithms and Architectures
Country/TerritoryCanada
CityWinnipeg, MAN.
Period10/08/0213/08/02

Keywords

  • Distributed
  • Load-balancing
  • Wait-free

Fingerprint

Dive into the research topics of 'Work dealing'. Together they form a unique fingerprint.

Cite this