Research Of Cloud Service Composition Based On Wolf Pack Algorithm

As a new type of service model,cloud services have developed rapidly in recent years due to their features of high reliability,low cost,and on-demand services.the trend of explosive growth.However,a single cloud service can no longer meet the increasingly complex and diversified service needs of users,and it needs to provide services to users after the combination of cloud services.Under such a background,how to quickly and effectively combine a large number of candidate cloud services according to user requirements and optimize the current problems that need to be solved for the optimal combination services.Most of the current researches on cloud service composition only use the general QoS attributes of cloud services as evaluation indicators,and do not consider factors such as CPU,memory and bandwidth that affect system load balancing.Most of the cloud service composition models are solved using intelligent algorithms.When these algorithms are solved,they may fall into a local optimum,and affecting the accuracy of the cloud service composition.Therefore,it is very important to quickly and accurately find cloud services that can meet user requirements and have optimal QoS values from massive cloud services to perform service composition,which can reduce costs and improve cloud service quality and user satisfaction.First,consider the load balance factors such as CPU usage,Bandwidth occupancy and memory occupancy,and the impact of common QoS attributes of cloud services such as time,cost,reliability,availability,and reputation on cloud services,and give quantitative expressions for each QoS attribute.A QoS evaluation model for cloud services was established.Then,after analyzing the detailed flow of cloud service composition and the four basic structures among cloud services in the cloud service composition,the global optimal cloud service composition QoS calculation model is given.Secondly,based on the above-mentioned model,the wolves algorithm is used to optimize the cloud service composition.Solve the position of the individual wolves by integer coding.Correspond to the number of each cloud service in the subtask candidate set in each cloud service composition path,solve the QoS optimal cloud service combination execution path,and then use broken paper to restore the specific Application scenarios and design simulation experiments verify the effectiveness and efficiency of the Wolves algorithm in solving the cloud service composition model.Finally,some improvements in the initialization,wandering behavior and summoning behavior of Wolf Pack algorithm are proposed.Then the improved Wolf Pack algorithm is used to solve the cloud service composition model.The same application scenario is used to design the simulation experiment.Compared with the results of the wolf pack algorithm and the adaptive particle swarm algorithm,the effectiveness of the improved algorithm is verified.