Consensus algorithms are considered the building blocks for building fault tolerant distributed systems, including distributed collaborative systems.
In the context of the consensus problem, two types of distributed systems should be distinguished : synchronized and non-synchronized. Resolving the consensus problem in the first is not a big problem, but in the second, the task is impossible, since it is not possible to distinguish between the failure of processes and between the failure or slow messages arrive. In this paper we present three of the most famous algorithms to solve consensus problem in asynchronous distributed systems: the Chandra and S.Toueg algorithm, the PAXOS algorithm, the A. Mostefaoui and M. Raynal algorithm. we compare between them to determine the most effective algorithm in case of increasing Number of processes (scalable distributed systems). We are using three basic standards for performance measure standards
Al-Baath University Journal.
2018.
A Reference Study of consensus Algorithms in Extended Distributed Systems - file Number 40.
Students
Teachers
Graduates
Faculties