### 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 language | English (US) |
---|---|

Title of host publication | Distributed Algorithms - 11th International Workshop, WDAG 1997, Proceedings |

Editors | Philippas Tsigas, Marios Mavronicolas |

Publisher | Springer Verlag |

Pages | 96-110 |

Number of pages | 15 |

ISBN (Print) | 3540635750, 9783540635758 |

State | Published - Jan 1 1997 |

Externally published | Yes |

Event | 11th International Workshop on Distributed Algorithms, WDAG 1997 - Saarbrucken, Germany Duration: Sep 24 1997 → Sep 26 1997 |

### Publication series

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

Volume | 1320 |

ISSN (Print) | 0302-9743 |

ISSN (Electronic) | 1611-3349 |

### Conference

Conference | 11th International Workshop on Distributed Algorithms, WDAG 1997 |
---|---|

Country | Germany |

City | Saarbrucken |

Period | 9/24/97 → 9/26/97 |

### Fingerprint

### ASJC Scopus subject areas

- Theoretical Computer Science
- Computer Science(all)

### Cite this

*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.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

*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.

}

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 -