Quick atomic broadcast

Piotr Berman, Anupam A. Bharali

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

    4 Citations (Scopus)

    Abstract

    In Atomic Broadcast protocols, the participating processors may initiate the broadcast of a message at any time; the goal is to make all non-faulty processors deliver the same set of messages in the same order. Our objective in this paper is to assure that the delivery time (from the moment of initiation of a message to its last delivery by a correct processor) is proportional to the actual number of faults in a given run of the protocol. We study this problem in synchronous systems and two models of semi-synchronous systems in presence of various classes of processor failures. We offer a technique that allows to convert “early stopping” Distributed Consensus protocols to Atomic Broadcast protocols with the above property. The resulting Atomic Broadcast protocols are faster and use smaller messages than the protocols proposed so far.

    Original languageEnglish (US)
    Title of host publicationDistributed Algorithms - 7th International Workshop, WDAG 1993, Proceedings
    EditorsAndre Schipe
    PublisherSpringer Verlag
    Pages189-203
    Number of pages15
    ISBN (Print)9783540572718
    StatePublished - Jan 1 1993
    Event7th International Workshop on Distributed Algorithms, WDAG 1993 - Lausanne, Switzerland
    Duration: Sep 27 1993Sep 29 1993

    Publication series

    NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
    Volume725 LNCS
    ISSN (Print)0302-9743
    ISSN (Electronic)1611-3349

    Other

    Other7th International Workshop on Distributed Algorithms, WDAG 1993
    CountrySwitzerland
    CityLausanne
    Period9/27/939/29/93

    Fingerprint

    Broadcast
    Synchronous Systems
    Early Stopping
    Convert
    Fault
    Directly proportional
    Moment

    All Science Journal Classification (ASJC) codes

    • Theoretical Computer Science
    • Computer Science(all)

    Cite this

    Berman, P., & Bharali, A. A. (1993). Quick atomic broadcast. In A. Schipe (Ed.), Distributed Algorithms - 7th International Workshop, WDAG 1993, Proceedings (pp. 189-203). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 725 LNCS). Springer Verlag.
    Berman, Piotr ; Bharali, Anupam A. / Quick atomic broadcast. Distributed Algorithms - 7th International Workshop, WDAG 1993, Proceedings. editor / Andre Schipe. Springer Verlag, 1993. pp. 189-203 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
    @inproceedings{0111d29db9f5419984b0a4e916ec075c,
    title = "Quick atomic broadcast",
    abstract = "In Atomic Broadcast protocols, the participating processors may initiate the broadcast of a message at any time; the goal is to make all non-faulty processors deliver the same set of messages in the same order. Our objective in this paper is to assure that the delivery time (from the moment of initiation of a message to its last delivery by a correct processor) is proportional to the actual number of faults in a given run of the protocol. We study this problem in synchronous systems and two models of semi-synchronous systems in presence of various classes of processor failures. We offer a technique that allows to convert “early stopping” Distributed Consensus protocols to Atomic Broadcast protocols with the above property. The resulting Atomic Broadcast protocols are faster and use smaller messages than the protocols proposed so far.",
    author = "Piotr Berman and Bharali, {Anupam A.}",
    year = "1993",
    month = "1",
    day = "1",
    language = "English (US)",
    isbn = "9783540572718",
    series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
    publisher = "Springer Verlag",
    pages = "189--203",
    editor = "Andre Schipe",
    booktitle = "Distributed Algorithms - 7th International Workshop, WDAG 1993, Proceedings",
    address = "Germany",

    }

    Berman, P & Bharali, AA 1993, Quick atomic broadcast. in A Schipe (ed.), Distributed Algorithms - 7th International Workshop, WDAG 1993, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 725 LNCS, Springer Verlag, pp. 189-203, 7th International Workshop on Distributed Algorithms, WDAG 1993, Lausanne, Switzerland, 9/27/93.

    Quick atomic broadcast. / Berman, Piotr; Bharali, Anupam A.

    Distributed Algorithms - 7th International Workshop, WDAG 1993, Proceedings. ed. / Andre Schipe. Springer Verlag, 1993. p. 189-203 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 725 LNCS).

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

    TY - GEN

    T1 - Quick atomic broadcast

    AU - Berman, Piotr

    AU - Bharali, Anupam A.

    PY - 1993/1/1

    Y1 - 1993/1/1

    N2 - In Atomic Broadcast protocols, the participating processors may initiate the broadcast of a message at any time; the goal is to make all non-faulty processors deliver the same set of messages in the same order. Our objective in this paper is to assure that the delivery time (from the moment of initiation of a message to its last delivery by a correct processor) is proportional to the actual number of faults in a given run of the protocol. We study this problem in synchronous systems and two models of semi-synchronous systems in presence of various classes of processor failures. We offer a technique that allows to convert “early stopping” Distributed Consensus protocols to Atomic Broadcast protocols with the above property. The resulting Atomic Broadcast protocols are faster and use smaller messages than the protocols proposed so far.

    AB - In Atomic Broadcast protocols, the participating processors may initiate the broadcast of a message at any time; the goal is to make all non-faulty processors deliver the same set of messages in the same order. Our objective in this paper is to assure that the delivery time (from the moment of initiation of a message to its last delivery by a correct processor) is proportional to the actual number of faults in a given run of the protocol. We study this problem in synchronous systems and two models of semi-synchronous systems in presence of various classes of processor failures. We offer a technique that allows to convert “early stopping” Distributed Consensus protocols to Atomic Broadcast protocols with the above property. The resulting Atomic Broadcast protocols are faster and use smaller messages than the protocols proposed so far.

    UR - http://www.scopus.com/inward/record.url?scp=84988666402&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84988666402&partnerID=8YFLogxK

    M3 - Conference contribution

    SN - 9783540572718

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

    SP - 189

    EP - 203

    BT - Distributed Algorithms - 7th International Workshop, WDAG 1993, Proceedings

    A2 - Schipe, Andre

    PB - Springer Verlag

    ER -

    Berman P, Bharali AA. Quick atomic broadcast. In Schipe A, editor, Distributed Algorithms - 7th International Workshop, WDAG 1993, Proceedings. Springer Verlag. 1993. p. 189-203. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).