Research On Dynamic Demand Decomposition And Resource Bidirectional Matching In Open Environment

With the continuous development of web services,the traditional mode which satisfies the various requests of the user with a single service is time-consuming and labor-intensive,and the concept of service combination provides a new idea for solving this problem.Service composition is the process of integrating several atomic services with simple functions into integrated services with more complex functions through integrated encapsulation.Service composition can increase service value by improving service reusability.In the complex and changeable open environment,the demand changes with the change of the user's goal,the web service born or disappeared through its life span,all of them are not constant,a large number of users and services exist dynamically,how to analyze uses' dynamical requirements,then search and combine dynamical atomic services is an important content in the research of service composition.For the service composition that involves computing resource allocation,there are three kinds of roles: resource providers,service providers and users.The traditional service allocation often takes the interest demands of the service demand side into account only,and neglects the independent choice of the service supplier to the customer.Because of the limited number of service resources and the limitation of the cost,this way of distribution is not conducive for the development of the market.The cooperation between the supply and demand parties should be the result of two-way choice.At the same time,with the development of network resources and standardized management,more service providers choose to purchase the computing resourcesof virtual servers to provide services to users.How to realize the resource balance of virtual servers affects the efficiency of the environment and the utilization of resources.Under such a background,this paper mainly studies the dynamic decomposition of demand and resource allocation in service composition in open environment.The main contents are as follows:1.We propose a dynamic decomposition method of user requirements in open environment.In order to solve the problem about how to respond to these changes in the process of demand decomposition,a scheme of setting up a requirement decomposition target graph is designed,which takes the demand of users and the real-time performance of service state into account.2.We propose a stable service matching method based on both supply and demand sides.The resource allocation method considers the demander' satisfactory on web services,as well as supplier' cost-effectiveness of web services to make the matching scheme reach the global optimal.3.We propose a method considering the pricing of virtual server resource balance.It based on a time series based virtual server pricing method,so it can enables the service providers to achieve the equilibrium of virtual server resources under the maximization of revenue selection strategy.In this paper,a part of simulation experiments are designed,and the test results are analyzed and discussed,then we verified the effectiveness of the proposed algorithm.Finally,a service combination public platform prototype has been created to provide practical reference for automated service composition technology.