Web Service Composition Algorithms In Cloud Environment

With the continuous evolution of cloud computing and widespread deployment of cloud platforms,diverse business modules have been migrated to cloud platforms and published as web services on the Internet.Web service is always built with simple functionality,which cannot meet the complex user requirements.Thus it is essential to composite several different services.In the cloud environment,the large quantity of web services with interfaces defined variously along with a great number of complex and personalized service requests will bring great challenges to service composition.User requirements include functional service requirements that illustrate the desirable functionalities and non-functional service requirements that indicate the user's QoS demands.Considering the features of cloud environment,this thesis chooses automatic semantic web service composition model,based on which several service composition algorithms are proposed to deal with the user requirements.The main works in this thesis are as follows.(1)The semantic web service composition is formulated from the perspectives of parameter matching and service matching respectively.As such,the service composition process is decomposed to three stages: parameter expansion,service dependency graph construction,and backtracking for composited service.The layered expansion strategy as well as the non-layered expansion strategy are proposed for the iterative parameter expansion process,meanwhile the concept of virtual parameter is introduced to improve the efficiency of service composition.(2)To address the functional service requirements,a layered service composition algorithm and a non-layered service composition algorithm are proposed on the basis of the two parameter expansion strategies respectively.Both theoretical analysis and experimental results show that these algorithms can obtain a feasible composited service efficiently.Moreover,the layered expansion strategy can achieve better performance.(3)To address both functional and non-functional service requirements,a QoS aware service composition algorithm is proposed,which takes the QoS attribute into consideration in the service composition procedure.The performance of the algorithm is improved by adjusting parameter dependency dynamically and taking use of the layer states.Both theoretical analysis and experimental results show that this algorithm can obtain a composited service with the optimal QoS.In conclusion,this thesis investigates the web service composition problem in cloud environment.To meet the functional and non-functional service requirements,several service composition algorithms are proposed,and the validity and efficiency of these algorithms are proved by both theoretical analysis and experimental results.This work can provide a theoretical basis for extensive researches and applications of web service technologies.