Fast scalable deterministic consensus for crash failures

Bogdan S. Chlebus, Dariusz R. Kowalski, Michał Strojnowski

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

Abstract

We study communication complexity of consensus in synchronous message-passing systems with processes prone to crashes. The goal in the consensus problem is to have all the nonfaulty processes agree on a common value from among the input ones, after each process has been initialized with a binary input value. The system consists of n processes and it is assumed that at most t < n processes crash in an execution. A consensus algorithm that tolerates up to t failures is called fast when its time complexity is script O sign(t). All the previously known fast deterministic consensus solutions sent Ω(n2) bits in messages. We give a fast deterministic consensus algorithm that has processes send only script O sign(n log4 n) bits. In our solution, processes exchange messages according to topologies of overlay graphs that have suitable robustness and connectivity properties related to graph expansion.

Original languageEnglish (US)
Title of host publicationPODC'09 - Proceedings of the 2009 ACM Symposium on Principles of Distributed Computing
Pages111-120
Number of pages10
DOIs
StatePublished - Nov 9 2009
Externally publishedYes
Event2009 ACM Symposium on Principles of Distributed Computing, PODC'09 - Calgary, AB, Canada
Duration: Aug 10 2009Aug 12 2009

Publication series

NameProceedings of the Annual ACM Symposium on Principles of Distributed Computing

Conference

Conference2009 ACM Symposium on Principles of Distributed Computing, PODC'09
CountryCanada
CityCalgary, AB
Period8/10/098/12/09

Fingerprint

Message passing
Topology
Communication

Keywords

  • Bit communication complexity
  • Consensus
  • Fault tolerance
  • Graph expansion
  • Message passing
  • Scalability
  • Synchrony

ASJC Scopus subject areas

  • Software
  • Hardware and Architecture
  • Computer Networks and Communications

Cite this

Chlebus, B. S., Kowalski, D. R., & Strojnowski, M. (2009). Fast scalable deterministic consensus for crash failures. In PODC'09 - Proceedings of the 2009 ACM Symposium on Principles of Distributed Computing (pp. 111-120). [1582738] (Proceedings of the Annual ACM Symposium on Principles of Distributed Computing). https://doi.org/10.1145/1582716.1582738

Fast scalable deterministic consensus for crash failures. / Chlebus, Bogdan S.; Kowalski, Dariusz R.; Strojnowski, Michał.

PODC'09 - Proceedings of the 2009 ACM Symposium on Principles of Distributed Computing. 2009. p. 111-120 1582738 (Proceedings of the Annual ACM Symposium on Principles of Distributed Computing).

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

Chlebus, BS, Kowalski, DR & Strojnowski, M 2009, Fast scalable deterministic consensus for crash failures. in PODC'09 - Proceedings of the 2009 ACM Symposium on Principles of Distributed Computing., 1582738, Proceedings of the Annual ACM Symposium on Principles of Distributed Computing, pp. 111-120, 2009 ACM Symposium on Principles of Distributed Computing, PODC'09, Calgary, AB, Canada, 8/10/09. https://doi.org/10.1145/1582716.1582738
Chlebus BS, Kowalski DR, Strojnowski M. Fast scalable deterministic consensus for crash failures. In PODC'09 - Proceedings of the 2009 ACM Symposium on Principles of Distributed Computing. 2009. p. 111-120. 1582738. (Proceedings of the Annual ACM Symposium on Principles of Distributed Computing). https://doi.org/10.1145/1582716.1582738
Chlebus, Bogdan S. ; Kowalski, Dariusz R. ; Strojnowski, Michał. / Fast scalable deterministic consensus for crash failures. PODC'09 - Proceedings of the 2009 ACM Symposium on Principles of Distributed Computing. 2009. pp. 111-120 (Proceedings of the Annual ACM Symposium on Principles of Distributed Computing).
@inproceedings{484d170a92da4e3796dbb3efebd91d22,
title = "Fast scalable deterministic consensus for crash failures",
abstract = "We study communication complexity of consensus in synchronous message-passing systems with processes prone to crashes. The goal in the consensus problem is to have all the nonfaulty processes agree on a common value from among the input ones, after each process has been initialized with a binary input value. The system consists of n processes and it is assumed that at most t < n processes crash in an execution. A consensus algorithm that tolerates up to t failures is called fast when its time complexity is script O sign(t). All the previously known fast deterministic consensus solutions sent Ω(n2) bits in messages. We give a fast deterministic consensus algorithm that has processes send only script O sign(n log4 n) bits. In our solution, processes exchange messages according to topologies of overlay graphs that have suitable robustness and connectivity properties related to graph expansion.",
keywords = "Bit communication complexity, Consensus, Fault tolerance, Graph expansion, Message passing, Scalability, Synchrony",
author = "Chlebus, {Bogdan S.} and Kowalski, {Dariusz R.} and Michał Strojnowski",
year = "2009",
month = "11",
day = "9",
doi = "10.1145/1582716.1582738",
language = "English (US)",
isbn = "9781605583969",
series = "Proceedings of the Annual ACM Symposium on Principles of Distributed Computing",
pages = "111--120",
booktitle = "PODC'09 - Proceedings of the 2009 ACM Symposium on Principles of Distributed Computing",

}

TY - GEN

T1 - Fast scalable deterministic consensus for crash failures

AU - Chlebus, Bogdan S.

AU - Kowalski, Dariusz R.

AU - Strojnowski, Michał

PY - 2009/11/9

Y1 - 2009/11/9

N2 - We study communication complexity of consensus in synchronous message-passing systems with processes prone to crashes. The goal in the consensus problem is to have all the nonfaulty processes agree on a common value from among the input ones, after each process has been initialized with a binary input value. The system consists of n processes and it is assumed that at most t < n processes crash in an execution. A consensus algorithm that tolerates up to t failures is called fast when its time complexity is script O sign(t). All the previously known fast deterministic consensus solutions sent Ω(n2) bits in messages. We give a fast deterministic consensus algorithm that has processes send only script O sign(n log4 n) bits. In our solution, processes exchange messages according to topologies of overlay graphs that have suitable robustness and connectivity properties related to graph expansion.

AB - We study communication complexity of consensus in synchronous message-passing systems with processes prone to crashes. The goal in the consensus problem is to have all the nonfaulty processes agree on a common value from among the input ones, after each process has been initialized with a binary input value. The system consists of n processes and it is assumed that at most t < n processes crash in an execution. A consensus algorithm that tolerates up to t failures is called fast when its time complexity is script O sign(t). All the previously known fast deterministic consensus solutions sent Ω(n2) bits in messages. We give a fast deterministic consensus algorithm that has processes send only script O sign(n log4 n) bits. In our solution, processes exchange messages according to topologies of overlay graphs that have suitable robustness and connectivity properties related to graph expansion.

KW - Bit communication complexity

KW - Consensus

KW - Fault tolerance

KW - Graph expansion

KW - Message passing

KW - Scalability

KW - Synchrony

UR - http://www.scopus.com/inward/record.url?scp=70350641373&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=70350641373&partnerID=8YFLogxK

U2 - 10.1145/1582716.1582738

DO - 10.1145/1582716.1582738

M3 - Conference contribution

AN - SCOPUS:70350641373

SN - 9781605583969

T3 - Proceedings of the Annual ACM Symposium on Principles of Distributed Computing

SP - 111

EP - 120

BT - PODC'09 - Proceedings of the 2009 ACM Symposium on Principles of Distributed Computing

ER -