Communication-efficient leader election in crash–recovery systems

• We implement Omega in crash–recovery partially synchronous distributed systems. • A first algorithm uses stable storage and is communication-efficient. • A second algorithm does not use stable storage and is near-communication-efficient. • The algorithms differ on the degree of agreement of unstab...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:The Journal of systems and software Ročník 84; číslo 12; s. 2186 - 2195
Hlavní autoři: Larrea, Mikel, Martín, Cristian, Soraluze, Iratxe
Médium: Journal Article
Jazyk:angličtina
Vydáno: New York Elsevier Inc 01.12.2011
Elsevier Sequoia S.A
Témata:
Algorithms
Communication
Communication-efficient algorithm
Computer programs
Consensus
Crashes
Crash–recovery
Distributed processing
Elections
Failure
Failure analysis
Fault-tolerant distributed computing
Information storage
Leader election
Messages
Omega failure detector
Receiving
Studies
Synchronous
Communication-efficient algorithm
Crash–recovery
Omega failure detector
Fault-tolerant distributed computing
Leader election
Consensus
ISSN:0164-1212, 1873-1228
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:• We implement Omega in crash–recovery partially synchronous distributed systems. • A first algorithm uses stable storage and is communication-efficient. • A second algorithm does not use stable storage and is near-communication-efficient. • The algorithms differ on the degree of agreement of unstable processes. This work addresses the leader election problem in partially synchronous distributed systems where processes can crash and recover. More precisely, it focuses on implementing the Omega failure detector class, which provides a leader election functionality, in the crash–recovery failure model. The concepts of communication efficiency and near-efficiency for an algorithm implementing Omega are defined. Depending on the use or not of stable storage, the property satisfied by unstable processes, i.e., those that crash and recover infinitely often, varies. Two algorithms implementing Omega are presented. In the first algorithm, which is communication-efficient and uses stable storage, eventually and permanently unstable processes agree on the leader with correct processes. In the second algorithm, which is near-communication-efficient and does not use stable storage, processes start their execution with no leader in order to avoid the disagreement among unstable processes, that will agree on the leader with correct processes after receiving a first message from the leader.
Bibliografie:SourceType-Scholarly Journals-1
ObjectType-Feature-1
content type line 14
ObjectType-Article-2
content type line 23
ISSN:0164-1212
1873-1228
DOI:10.1016/j.jss.2011.06.019