Maximum recovery time analysis of DRP for realtime industrial networks

In network control systems based on industrial Ethernet protocol, the availability and reliability of the system are indispensable. Redundancy is the most effective way to guarantee availability and reliability in a system. This paper introduces a distributed redundancy protocol DRP with a ring network architecture which can detect a single failure in the network and recover it within a deterministic time limit. The performance of DRP is analyzed by focusing on the maximum recovery time of the protocol. A time model to get the maximum recovery time is established. Tests with a prototype system and simulation results prove the effectiveness of the time model.