A Survey on Fault-Tolerance in Distributed Network Systems

In this paper, we give a survey on fault tolerant issue in distributed systems. More specially speaking, we talk about one important and basic component called failure detection, which is to detect the failure of the process quickly and accurately. Thus, a good failure detection method will avoid the further system lost due to process crash. This survey provides the related research results and also explored the future directions about failure detection, and it is a good reference for researcher on this topic.