Research On Service Composition Optimization Problem Based On Quotient Space Granulating

Service computing,as a new paradigm of computing,develops rapidly with its features of service-oriented building blocks and supporting fast,low-cost and convenient combination of distributed applications in heterogeneous environments.The core idea of service computing is to reuse instead of re-developing new enterprise applications,so some web services enable fast,flexible,seamless integration and collaboration among different business applications distributed within or across enterprise boundaries.However,with the development of SOAP and XML technologies,more and more network resources are released and used in the form of services.The number of services with the same function increases rapidly,combining with diversified,dynamic and complicated user needs so that service portfolio shows an exponential growth trend,service composition problems have gradually evolved into a NP-complete problem.In particular,with the complication of user needs and the single service fails to meet the needs of users,services composition is appeared and their scale of computing have grown exponentially.In the study of service composition optimization problem,this article is also referred to as service composition problem.Commonly used methods include intelligent optimization algorithm and mathematical optimization technology.Where intelligent optimization algorithms include Particle Swarm Optimization(PSO),Genetic Algorithm(GA),Ant Colony Algorithm(ACO),Artificial bee colony algorithm(ABC)and the Fruit fly Optimization Algorithm(FOA),mathematical optimization techniques include linear programming,dynamic programming,graph algorithms and integer programming.The advantage of intelligent optimization algorithm is that it takes less time to obtain user satisfaction solution,but it often cannot get the global optimal solution,while the mathematical optimization technology can obtain the global optimal solution,but its computation time increases significantly with the increase of service scale,usually exponential changes.Based on this,some researchers propose a decomposition scheme of service composition optimization problem,that is,decompose the original composition optimization problem into several simple sub-composition optimization problems,and then use the optimal solution of sub-problems to obtain a feasible solution to the original problem.However,in the process of problem decomposition and result summary,this model often results in the loss of accuracy and the global optimal solution cannot be guaranteed.Based on this,in order to solve the problem of large-scale service composition optimization,this chapter introduces the idea of granulating based on quotient-space theory to solve the service composition optimization problem and the service composition optimization of QoS constraint problem.In the existing research methods,multi-default services are independent of each other.However,in the real world,due to the benefits among service providers,there are many correlations and connections between services and services.For example,in an online shopping process,there is free shipping.In order to solve the problem of large-scale service composition and quality correlation service composition quickly and effectively,this chapter makes use of the idea of quotient space granulating in the two aspects of task granulating and correlation granulating under single and multiple attributes solution to service composition optimization problem.The main tasks include:1 this paper introduces the introduction and development of service composition optimization problems,especially introduces the quality correlation aware service composition optimization problem,and elaborates the service composition optimization model of service composition optimization model and quality correlation perception in detail.Then,based on the quality correlation aware service composition optimization model,a service composition optimization method based on task granulating and correlation granulating is proposed.2 In the stage of task granulating,the task granularity is processed through the membership relations between tasks,the service composition problem is decomposed into multiple sub-service composition problems by using the task granulating results,and the result of sub-service composition problem is gradually approached.Based on this,the task-granulating algorithm TgA is proposed and its feasibility and accuracy are verified theoretically.Finally,a large number of simulation experiments are carried out to verify the performance of task-granulating algorithm.3 Constraint granulation methods is based on correlation aggregation,that is,through the granulating process between the quality correlation relations,the quality correlation granularity becomes larger,and the number of times of quality correlation query is gradually reduced,and finally the query time is reduced so as to achieve the optimal query time.Through theoretical analysis,the idea of using granulation can effectively solve the service composition problem of large-scale and even oversized service composition and correlation perception.Based on this,this paper proposes a correlation granulating algorithm Q2C and a quality-aware service composition optimization algorithm Q2CO,and verifies the performance of the algorithm through a large number of simulation experiments.