The Research On Cloud Computing Oriented Service Selection

With the development of cloud computing technology,the user can obtain the resources they need through Internet.The system responses users5 requests in the way of using cloud service.However,as to the same request,have more than one service meet the functional demands,which meet the different non-functional demands(QoS).Thus,how to select the trustworthy,available and dependable cloud services become very important research.At the same time,Users' requirements are always numerous and complicated.It is difficult to select proper atomic services to meet users'requirements,so need use the technology of service composition.Therefore,it is a challenging research to select suitable values of atomic considering the global QoS optimal.Aim at these two problems,this paper focused on follow work:First of all,in order to improve the efficiency of cloud service selection and guarantee choose the trustworthy,available and dependable cloud service.Therefore,in this article we propose a novel model of cloud service selection based on trust trend.This model calculates the final trust value through two parts:initial trust value and trust trend value(TTV).Initial trust value is calculated through Bayes theorem.Trust trend value is calculated based on least squares linear regression,trust trend value aims to illustrate the trust trend of changes in a given period.We aggregate the two values to confirm the final trust value.Then,obtain the objective QoS value according to the QoS quantitative model of could services.At the same time,the measuring strategy of trust relationship among cloud services based on information entropy was designed.The experimental result shows that the method can reflect changes in trust cloud services,and enhanced the predictive ability,effectively improve the success rate of cloud service selection.Secondly,it is a challenging research to select suitable values of atomic considering the global QoS optimal,and the existing particle swarm service selection algorithm easy to get into local optimum and premature convergence problem.we propose an adaptive chaotic particle swarm service selection algorithm(ACPSO),in which the inertia weight of the particle was adjusted adaptively based on the premature convergence degree of the swarm and the fitness of the particle,the diversity of inertia weight makes a compromise between the global convergence and convergence speed,so it can effectively alleviate the problem of premature convergence.The research of service selection under the cloud environment has not yet been saturated,the further study of cloud service selection can promote further popularization and development of cloud computing.