Find in Library

Failure detection is one of the basic functions of building a reliable disaster recovery backup system. Aiming at the application-level disaster recovery backup failure detection problem, this paper analyzes the remote disaster recovery center architecture and failure detection hierarchy, and predicts the arrival time of cross-domain heartbeat information through the back propagation neural network based on particle swarm optimization (PSO-BP). When the actual timeout is reached, the active Auxiliary Detection (AD) is used to improve the correctness of failure detection, and finally the effectiveness of method PSO-BP-AD is verified through simulation.