融合网络环境下快速可靠的服务组合容错方法

Fast and Reliable Fault-Tolerance Approach for Service Composition in Integration Networks

针对传统容错方法在融合网络环境下服务组合的低效性,提出了一种快速可靠的服务组合容错方法.该方法首先采用模糊逻辑对服务的临时性故障进行服务重试;然后采用多属性决策理论对服务的永久性故障进行服务复制;最后,通过改进的粒子群算法对永久性故障进行服务补偿.基于真实数据集的实验结果表明,所提方法在故障排除率、故障处理时间与组合最优度方面均优于其他方法.

Traditional fault-tolerance approaches often result in low efficiency of service composition in integration networks.In this paper,a fast and reliable fault-tolerance approach is proposed for service composition in integration networks.This approach firstly adopts fuzzy logic to perform service retry when the transient faults of service occur.And then multi-attribute decision-making theory is employed to carry out service replicate when the permanent faults of service occur.Finally,an improved particle swarm optimization algorithm is used to implement service compensation when the permanent faults of service arise.The experimental results based on real data sets show that the proposed approach is superior to other approaches in terms of fault handling rate,fault handling time,and composition optimization.