Coordinated cooperative task computing using crash-prone processors with unreliable multicast

Seda Davtyan, Roberto De Prisco, Chryssis Georgiou, Theophanis Hadjistasi, Alexander Allister Schwarzmann

Research output: Contribution to journalArticle

Abstract

This paper presents a new message-passing algorithm, called Do-UM, for distributed cooperative task computing in synchronous settings where processors may crash, and where any multicasts (or broadcasts) performed by crashing processors are unreliable. We specify the algorithm, prove its correctness and analyse its complexity. We show that its worst case available processor steps is S=Θt+n [Formula presented] +f(n−f) and that the number of messages sent is less than n2t+ [Formula presented], where n is the number of processors, t is the number of tasks to be executed and f is the number of failures. To assess the performance of the algorithm in practical scenarios, we perform an experimental evaluation on a planetary-scale distributed platform. This also allows us to compare our algorithm with the currently best algorithm that is, however, explicitly designed to use reliable multicast; the results suggest that our algorithm does not lose much efficiency in order to cope with unreliable multicast.

Original languageEnglish (US)
Pages (from-to)272-285
Number of pages14
JournalJournal of Parallel and Distributed Computing
Volume109
DOIs
StatePublished - Nov 1 2017

Fingerprint

Crash
Multicast
Computing
Message-passing Algorithms
Experimental Evaluation
Broadcast
Message passing
Correctness
Scenarios

Keywords

  • Crash faults
  • Fault-tolerant distributed algorithms
  • Task computing
  • Unreliable multicast

ASJC Scopus subject areas

  • Software
  • Theoretical Computer Science
  • Hardware and Architecture
  • Computer Networks and Communications
  • Artificial Intelligence

Cite this

Coordinated cooperative task computing using crash-prone processors with unreliable multicast. / Davtyan, Seda; De Prisco, Roberto; Georgiou, Chryssis; Hadjistasi, Theophanis; Schwarzmann, Alexander Allister.

In: Journal of Parallel and Distributed Computing, Vol. 109, 01.11.2017, p. 272-285.

Research output: Contribution to journalArticle

Davtyan, Seda ; De Prisco, Roberto ; Georgiou, Chryssis ; Hadjistasi, Theophanis ; Schwarzmann, Alexander Allister. / Coordinated cooperative task computing using crash-prone processors with unreliable multicast. In: Journal of Parallel and Distributed Computing. 2017 ; Vol. 109. pp. 272-285.
@article{455021300ed2455a8c7ec5d1c472039b,
title = "Coordinated cooperative task computing using crash-prone processors with unreliable multicast",
abstract = "This paper presents a new message-passing algorithm, called Do-UM, for distributed cooperative task computing in synchronous settings where processors may crash, and where any multicasts (or broadcasts) performed by crashing processors are unreliable. We specify the algorithm, prove its correctness and analyse its complexity. We show that its worst case available processor steps is S=Θt+n [Formula presented] +f(n−f) and that the number of messages sent is less than n2t+ [Formula presented], where n is the number of processors, t is the number of tasks to be executed and f is the number of failures. To assess the performance of the algorithm in practical scenarios, we perform an experimental evaluation on a planetary-scale distributed platform. This also allows us to compare our algorithm with the currently best algorithm that is, however, explicitly designed to use reliable multicast; the results suggest that our algorithm does not lose much efficiency in order to cope with unreliable multicast.",
keywords = "Crash faults, Fault-tolerant distributed algorithms, Task computing, Unreliable multicast",
author = "Seda Davtyan and {De Prisco}, Roberto and Chryssis Georgiou and Theophanis Hadjistasi and Schwarzmann, {Alexander Allister}",
year = "2017",
month = "11",
day = "1",
doi = "10.1016/j.jpdc.2017.06.013",
language = "English (US)",
volume = "109",
pages = "272--285",
journal = "Journal of Parallel and Distributed Computing",
issn = "0743-7315",
publisher = "Academic Press Inc.",

}

TY - JOUR

T1 - Coordinated cooperative task computing using crash-prone processors with unreliable multicast

AU - Davtyan, Seda

AU - De Prisco, Roberto

AU - Georgiou, Chryssis

AU - Hadjistasi, Theophanis

AU - Schwarzmann, Alexander Allister

PY - 2017/11/1

Y1 - 2017/11/1

N2 - This paper presents a new message-passing algorithm, called Do-UM, for distributed cooperative task computing in synchronous settings where processors may crash, and where any multicasts (or broadcasts) performed by crashing processors are unreliable. We specify the algorithm, prove its correctness and analyse its complexity. We show that its worst case available processor steps is S=Θt+n [Formula presented] +f(n−f) and that the number of messages sent is less than n2t+ [Formula presented], where n is the number of processors, t is the number of tasks to be executed and f is the number of failures. To assess the performance of the algorithm in practical scenarios, we perform an experimental evaluation on a planetary-scale distributed platform. This also allows us to compare our algorithm with the currently best algorithm that is, however, explicitly designed to use reliable multicast; the results suggest that our algorithm does not lose much efficiency in order to cope with unreliable multicast.

AB - This paper presents a new message-passing algorithm, called Do-UM, for distributed cooperative task computing in synchronous settings where processors may crash, and where any multicasts (or broadcasts) performed by crashing processors are unreliable. We specify the algorithm, prove its correctness and analyse its complexity. We show that its worst case available processor steps is S=Θt+n [Formula presented] +f(n−f) and that the number of messages sent is less than n2t+ [Formula presented], where n is the number of processors, t is the number of tasks to be executed and f is the number of failures. To assess the performance of the algorithm in practical scenarios, we perform an experimental evaluation on a planetary-scale distributed platform. This also allows us to compare our algorithm with the currently best algorithm that is, however, explicitly designed to use reliable multicast; the results suggest that our algorithm does not lose much efficiency in order to cope with unreliable multicast.

KW - Crash faults

KW - Fault-tolerant distributed algorithms

KW - Task computing

KW - Unreliable multicast

UR - http://www.scopus.com/inward/record.url?scp=85024405331&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85024405331&partnerID=8YFLogxK

U2 - 10.1016/j.jpdc.2017.06.013

DO - 10.1016/j.jpdc.2017.06.013

M3 - Article

AN - SCOPUS:85024405331

VL - 109

SP - 272

EP - 285

JO - Journal of Parallel and Distributed Computing

JF - Journal of Parallel and Distributed Computing

SN - 0743-7315

ER -