Research On Modeling Description And Service Compo Sition Optimization Of Enterprise Cloud Services Based On SaaS

With the development and application of the cloud computing technology,delivering applications and basic network resources to the user via the internet was becoming a mainstream consumption mode.Especially for medium and small-sized enterprise users,cloud services has become a major way to deploy enterprise software.However,due to the immateriality,the integration and the one-shot-principle referring to service products,it is more complicated to describe a service product than describing a traditional product.In addition to modeling service systems,software designers also need to model the infrastructure deployed by the system.This thesis analyzes and studies the modeling and description of enterprise cloud service system.Cloud computing can realize the integration and optimization of cyber source,and services reusing through service composition.Meanwhile,service composition is a kind of high efficiency method to get composite service.Cloud computing and service composition is mutual promotion.On the one hand,the resource integration capability of cloud computing can provide a number of services for service composition;on the other hand,the service composition technology is essential for cloud computing to realize the complex business process.In the face of the increasingly complex needs of users from different areas,cloud computing needs more extensive application and service composition requires more efficient.In order to enable enterprise users to customize the required services in the cloud platform.On the basis of enterprise cloud service model,it has important research value to provide a optimized combination method of Web service for enterprise cloud platform providers.The main works of this thesis are as follows:(1)The enterprise cloud service architecture is proposed,which is described in detail from the following five aspects: the types,characteristics,deployment,network foundation and service process.At the same time,based on the extension of unified modeling language,the modeling primitives for enterprise cloud services were designed and the equivalent element model of enterprise cloud services was established.The object constraint language guarantees the accuracy of the domain model.Finally,a cloud service model of enterprise resource planning based on the cloud service architecture was established.The model instance was described by building several visual diagram based on unified modeling language.The results verify the validity of the proposed modeling method and the consistency of the model.(2)A service selection method based on genetic algorithm is proposed.Firstly,presented the formal description of the non-functional attributes of the services.Then the non-functional attributes of the services were normalized by the aggregate function.At last,the rough set theory was introduced,organism combines rough set pretreatment with genetic algorithm to optimize the initial population and speed up the convergence.Meanwhile,in the iterative process of genetic algorithm introduced reserved population,which has a good solution to the problem of local optimum.