Waiting in concurrent algorithms

Gadi Taubenfeld

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Between the two extremes, lock-based algorithms, which involve “a lot of waiting”, and wait-free algorithms, which are “free of locking and waiting”, there is an interesting spectrum of different levels of waiting. This unexplored spectrum is formally defined and its properties are investigated. New progress conditions, called k-waiting, for k ≥ 0, which are intended to capture the “amount of waiting” of processes in asynchronous concurrent algorithms, are introduced. To illustrate the utility of the new conditions, they are used to derive new lower and upper bounds, and impossibility results for well-known basic problems such as consensus, election, renaming and mutual exclusion. Furthermore, the relation between waiting and fairness is explored.

Original languageEnglish
Title of host publicationNetworked Systems - 4th International Conference, NETYS 2016, Revised Selected Papers
EditorsCarole Delporte -Gallet, Parosh Aziz Abdulla
PublisherSpringer Verlag
Pages345-360
Number of pages16
ISBN (Print)9783319461397
DOIs
StatePublished - 2016
Externally publishedYes
Event4th International Conference on Networked Systems, NETYS 2016 - Marrakech, Morocco
Duration: 18 May 201620 May 2016

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume9944 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference4th International Conference on Networked Systems, NETYS 2016
Country/TerritoryMorocco
CityMarrakech
Period18/05/1620/05/16

Bibliographical note

Publisher Copyright:
© Springer International Publishing AG 2016.

Keywords

  • Consensus
  • Election
  • Enabled process
  • Enabling step
  • K-waiting
  • Locks
  • Mutual exclusion
  • Renaming
  • Synchronization
  • Wait-freedom

Fingerprint

Dive into the research topics of 'Waiting in concurrent algorithms'. Together they form a unique fingerprint.

Cite this