Mitigation strategies with feedback against cascading failures

Abstract

Because of the complexity and randomness of cascading failures, it is difficult to eliminate cascading failures and establish a robust network. However, we can sacrifice some unnecessary services to ensure the critical services of the network surviving under the attack. To reduce the impact of cascading failures on network robustness, we introduce and investigate a mitigation strategy with feedback against cascading failures. The mitigation strategy considers the flow dynamic, where the traffic exchanged between a pair of nodes can be adaptively adjusted depending on the changes of the shortest path length and the shortest path number between them. The simulations show that the mitigation strategy can effectively suppress the propagation of cascade. Particularly, there is an optimal feedback coefficient where the network is robust and the network traffic loss is minimal. It is shown to be more effective for suppressing the propagation of the cascade than the recent proposed strategy of mitigation, and under different cascading failure models, it can still suppress the propagation of the cascade well.