Performing tasks on restartable Message–Passing processors

Bogdan S. Chlebus, Roberto De Prisco, Alex A. Shvartsman

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

This work presents new algorithms for the “Do-All” problem that consists of performing t tasks reliably in a message-passing synchronous system of p fault-prone processors. The algorithms are based on an aggressive coordination paradigm in which multiple coordinators may be active as the result of failures. The first algorithm is tolerant of/< p stop-failures and it does not allow restarts. It has the available processor steps complexity S = 0((t+ plogp/log log p)–log f) and the message complexity M = 0(t + plogp/loglogp +/• p). Unlike prior solutions, our algorithm uses redundant broadcasts when encountering failures and, for large f, it has better S complexity. This algorithm is used as the basis for another algorithm which tolerates any pattern of stop-failures and restarts. This new algorithm is the first solution for the Do-All problem that efficiently deals with processor restarts. Its available processor steps complexity is 5 = 0((t+ p log p + f) -min{log p,log f}), and its message complexity is M = 0(t + p- logp +/-p), where/is the number of failures.

Original languageEnglish (US)
Title of host publicationDistributed Algorithms - 11th International Workshop, WDAG 1997, Proceedings
EditorsPhilippas Tsigas, Marios Mavronicolas
PublisherSpringer Verlag
Pages96-110
Number of pages15
ISBN (Print)3540635750, 9783540635758
StatePublished - Jan 1 1997
Externally publishedYes
Event11th International Workshop on Distributed Algorithms, WDAG 1997 - Saarbrucken, Germany
Duration: Sep 24 1997Sep 26 1997

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume1320
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference11th International Workshop on Distributed Algorithms, WDAG 1997
CountryGermany
CitySaarbrucken
Period9/24/979/26/97

Fingerprint

Restart
Message Complexity
Synchronous Systems
Message passing
Message Passing
Broadcast
Fault
Paradigm

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Chlebus, B. S., De Prisco, R., & Shvartsman, A. A. (1997). Performing tasks on restartable Message–Passing processors. In P. Tsigas, & M. Mavronicolas (Eds.), Distributed Algorithms - 11th International Workshop, WDAG 1997, Proceedings (pp. 96-110). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 1320). Springer Verlag.

Performing tasks on restartable Message–Passing processors. / Chlebus, Bogdan S.; De Prisco, Roberto; Shvartsman, Alex A.

Distributed Algorithms - 11th International Workshop, WDAG 1997, Proceedings. ed. / Philippas Tsigas; Marios Mavronicolas. Springer Verlag, 1997. p. 96-110 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 1320).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Chlebus, BS, De Prisco, R & Shvartsman, AA 1997, Performing tasks on restartable Message–Passing processors. in P Tsigas & M Mavronicolas (eds), Distributed Algorithms - 11th International Workshop, WDAG 1997, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 1320, Springer Verlag, pp. 96-110, 11th International Workshop on Distributed Algorithms, WDAG 1997, Saarbrucken, Germany, 9/24/97.
Chlebus BS, De Prisco R, Shvartsman AA. Performing tasks on restartable Message–Passing processors. In Tsigas P, Mavronicolas M, editors, Distributed Algorithms - 11th International Workshop, WDAG 1997, Proceedings. Springer Verlag. 1997. p. 96-110. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
Chlebus, Bogdan S. ; De Prisco, Roberto ; Shvartsman, Alex A. / Performing tasks on restartable Message–Passing processors. Distributed Algorithms - 11th International Workshop, WDAG 1997, Proceedings. editor / Philippas Tsigas ; Marios Mavronicolas. Springer Verlag, 1997. pp. 96-110 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{33d7492f90a74d77bc2032bf5479fa49,
title = "Performing tasks on restartable Message–Passing processors",
abstract = "This work presents new algorithms for the “Do-All” problem that consists of performing t tasks reliably in a message-passing synchronous system of p fault-prone processors. The algorithms are based on an aggressive coordination paradigm in which multiple coordinators may be active as the result of failures. The first algorithm is tolerant of/< p stop-failures and it does not allow restarts. It has the available processor steps complexity S = 0((t+ plogp/log log p)–log f) and the message complexity M = 0(t + plogp/loglogp +/• p). Unlike prior solutions, our algorithm uses redundant broadcasts when encountering failures and, for large f, it has better S complexity. This algorithm is used as the basis for another algorithm which tolerates any pattern of stop-failures and restarts. This new algorithm is the first solution for the Do-All problem that efficiently deals with processor restarts. Its available processor steps complexity is 5 = 0((t+ p log p + f) -min{log p,log f}), and its message complexity is M = 0(t + p- logp +/-p), where/is the number of failures.",
author = "Chlebus, {Bogdan S.} and {De Prisco}, Roberto and Shvartsman, {Alex A.}",
year = "1997",
month = "1",
day = "1",
language = "English (US)",
isbn = "3540635750",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "96--110",
editor = "Philippas Tsigas and Marios Mavronicolas",
booktitle = "Distributed Algorithms - 11th International Workshop, WDAG 1997, Proceedings",

}

TY - GEN

T1 - Performing tasks on restartable Message–Passing processors

AU - Chlebus, Bogdan S.

AU - De Prisco, Roberto

AU - Shvartsman, Alex A.

PY - 1997/1/1

Y1 - 1997/1/1

N2 - This work presents new algorithms for the “Do-All” problem that consists of performing t tasks reliably in a message-passing synchronous system of p fault-prone processors. The algorithms are based on an aggressive coordination paradigm in which multiple coordinators may be active as the result of failures. The first algorithm is tolerant of/< p stop-failures and it does not allow restarts. It has the available processor steps complexity S = 0((t+ plogp/log log p)–log f) and the message complexity M = 0(t + plogp/loglogp +/• p). Unlike prior solutions, our algorithm uses redundant broadcasts when encountering failures and, for large f, it has better S complexity. This algorithm is used as the basis for another algorithm which tolerates any pattern of stop-failures and restarts. This new algorithm is the first solution for the Do-All problem that efficiently deals with processor restarts. Its available processor steps complexity is 5 = 0((t+ p log p + f) -min{log p,log f}), and its message complexity is M = 0(t + p- logp +/-p), where/is the number of failures.

AB - This work presents new algorithms for the “Do-All” problem that consists of performing t tasks reliably in a message-passing synchronous system of p fault-prone processors. The algorithms are based on an aggressive coordination paradigm in which multiple coordinators may be active as the result of failures. The first algorithm is tolerant of/< p stop-failures and it does not allow restarts. It has the available processor steps complexity S = 0((t+ plogp/log log p)–log f) and the message complexity M = 0(t + plogp/loglogp +/• p). Unlike prior solutions, our algorithm uses redundant broadcasts when encountering failures and, for large f, it has better S complexity. This algorithm is used as the basis for another algorithm which tolerates any pattern of stop-failures and restarts. This new algorithm is the first solution for the Do-All problem that efficiently deals with processor restarts. Its available processor steps complexity is 5 = 0((t+ p log p + f) -min{log p,log f}), and its message complexity is M = 0(t + p- logp +/-p), where/is the number of failures.

UR - http://www.scopus.com/inward/record.url?scp=84949226040&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84949226040&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84949226040

SN - 3540635750

SN - 9783540635758

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

SP - 96

EP - 110

BT - Distributed Algorithms - 11th International Workshop, WDAG 1997, Proceedings

A2 - Tsigas, Philippas

A2 - Mavronicolas, Marios

PB - Springer Verlag

ER -