Research On The Self-organization Of Trusted Service Resource

The rapid development of mobile application and mobile Internet has promoted the wide application of mobile cloud computing.However,due to the large base of mobile users,centralized mobile cloud computing resource management system is prone to load bottlenecks,resulting in reduced quality of service.Based on the analysis of the concepts and technologies related to mobile cloud computing,the three key problems need to be solved in a cloud resource management system are proposed,including resource description,resource organization,resource scheduling.In order to provide users with reliable and high-quality cloud computing service,this thesis start research from the perspective of resource description and resource organization.The main work of this thesis is as follows.1.In order to get a unified description of heterogeneous,distributed,and complex large-scale resources in the cloud computing system.First,the resource are abstracted as computing resource,storage resource and bandwidth resource by analyzing the process of resource requests in 3 layer cloud services IaaS,PaaS and SaaS.Next,combines with the existing resource cataloguing techniques,this thesis proposes a unified identification method of multilayer service resource with semantics.Finally,the feasibility and validity of the unified identification method of multilayer service resource with semantics are verified by simulation experiments,which lays the foundation for self-organization of resource.2.In order to effectively organize the large-scale resources in the cloud computing system,and realize the dynamic addition of resource nodes and the rapid lookup discovery.First,this thesis proposes a self-organization method of service resource using the proposed method of unified identification method of multilayer service resource with semantics and the Pastry algorithm in structured P2 P network.Then,the method achieves the cloud resource node join,exit,update and user resource request routing.Finally,the simulation experiment proves that the proposed method based on Pastry technology has strong feasibility,and can greatly reduce the searching time when finding cloud resource and improve the efficiency of cloud resource management.From the credibility of service,for resource description and resource organization in cloud computing resource management system,this thesis proposes a unified identification method of multilayer service resource with semantics and a selforganization method of service resource based on the Pastry algorithm,which implementing different resource requirements from different users on a shared infrastructure for management and effectively reduces the discovery time of resource requests and laid the foundation of trustable cloud computing service to users.