Détecteurs de fautes
In distributed systems, failure detectors are used to monitor the processes and to reduce the risk of failures by detecting them before system crashes. Accuracy and completeness are the key attributes to measure the quality of failure detectors. Failure detectors are said to be unreliable because sometimes they suspect a correct process as a faulty process or they treat a faulty process as a correct process. In this paper various failure detector algorithms are discussed. A comprehensive study is presented based on properties, methodologies used, the applicability of systems, and outcomes of the failure detectors. The paper helps readers for the enhancement of knowledge about the basics of failure detectors and the different algorithms which are developed to solve the failure detection problems of distributed systems.