Microservice Composition Mechanism Based On QoS In Cyber-Physical Systems

With the digital transformation of enterprises,Cyber-Physical Systems(CPS)has been widely used in intelligent robots,smart factories,intelligent transportation,industrial monitoring and other industries.In order to adapt to the development of modern distributed systems,many CPS are deployed based on microservice architecture.With more and more service components accessing the CPS system,the services in CPS are more complex,massive and heterogeneous,and there are many uncertainties.At the same time,there are a lot of micro-services that need to be dynamically generated,which will inevitably lead to drastic fluctuations in service quality and affect the efficiency of service composition.In addition,high-concurrency application scenarios will also lead to the problem of Quality of Service(QoS)degradation of composite services due to overload of service instances.Therefore,this thesis studies the micro-service composition mechanism based on QoS in CPS,and proposes the optimal QoS service composition method based on Black Widow Optimization Algorithm(BWOA)and the service composition method based on load balance.The specific research work is as follows:1.Aiming at the low efficiency of micro-service composition caused by the large difference of service QoS attributes and massive service in CPS,an optimal QoS service composition method based on BWOA is proposed.According to the characteristics of CPS system,this method adopts microservice architecture,and divides services into application layer services,network layer services and physical layer services.According to diverse layers of services,the QoS model and the optimal QoS objective function are constructed respectively,and the objective function is taken as the corresponding adaptive function of each layer;Using Cauchy mutation and reverse learning strategy to optimize BWOA,and finally using the optimized BWOA to get the optimal QoS service combination.Compared with other heuristic algorithms in diverse QoS models and diverse services,the experimental results show that this method can effectively improve the efficiency and quality of service composition.2.In CPS open environment,when swarm intelligence algorithm is used for service selection,high QoS services are preferred.When the load of system service composition requests increases,high QoS services will be overloaded,which will greatly reduce the system service quality.To solve this problem,a service composition method based on load balance is proposed.This method constructs the overall load model,solves the adaptive function,reduces the service solution space through the dominance relation,and optimizes the heuristic algorithm for service composition.Compared with other methods in diverse requests,diverse service pools and diverse sub-services,the experimental results show that the proposed method can balance the system service load while maintaining high QoS.The research work shows that the method proposed in this thesis can effectively improve the QoS of microservice composition in CPS and ensure the load balance of microservice composition in high concurrency environment,which has great practical significance.