Leader Election in Distributed Systems with Crash Failures
نویسنده
چکیده
Leader election is an important problem in distributed computing. Garcia-Molina's Bully Algorithm is a classic solution to leader election in synchronous systems with crash failures. This paper shows that the Bully Algorithm can be easily adapted for use in asynchronous systems. First, we re-write the Bully Algorithm to use a failure detector, instead of explicit time-outs; this yields a modular solution to leader election in synchronous systems. Second, we show that minor modi cations to that algorithm yield a simple and e cient solution to leader election in asynchronous systems with crash failures. We point out a aw in Garcia-Molina's speci cation of leader election in asynchronous systems, propose a revised speci cation, and show that the modi ed Bully Algorithm satis es this speci cation.
منابع مشابه
Leader Election in Asynchronous Distributed Systems
In a previous paper, Garcia-Molina speci es the leader election problem for synchronous and asynchronous distributed systems with crash and link failures and gives an elegant algorithm for each type of system. This paper points out a aw in GarciaMolina's speci cation of leader election in asynchronous systems and proposes a new speci cation.
متن کاملStabilizing leader election in partial synchronous systems with crash failures
This article deals with stabilization and fault-tolerance. We consider two types of stabilization: the selfand the pseudostabilization. Our goal is to implement the selfand/or pseudostabilizing leader election in systems with process crashes, weak reliability, and synchrony assumptions. We try to propose, when it is possible, communication-efficient implementations. Our approach allows to obtai...
متن کاملRobust Stabilizing Leader Election
In this paper, we mix two well-known approaches of the fault-tolerance: robustness and stabilization. Robustnessis the aptitude of an algorithm to withstand permanent failures such as process crashes. The stabilization is a generaltechnique to design algorithms tolerating transient failures. Using these two approaches, we propose algorithms thattolerate both transient and crash fail...
متن کاملDesigning and Evaluating Fault-tolerant Leader Election Algorithms
Fault-tolerant leader election is a basic building block for dependable distributed computing, since it allows coordinating decisions among replicas such that they remain consistent. Indeed, several fault-tolerant agreement protocols rely on an eventual leader election service. This problem has been initially studied in crash-prone systems, and more recently in other failure scenarios, e.g., cr...
متن کاملAlgebraic Speciication of Dynamic Leader Election Protocols in Broadcast Networks
The problem of leader election in distributed systems is considered. Components communicate by means of buuered broadcasting as opposed to usual point-to-point communication. In this paper three leader election protocols of increasing maturity are speciied. We start with a simple leader election protocol, where an initial leader is present. In the second protocol this assumption is dropped. Eve...
متن کامل