Research Of Resource Provisioning Technique In Telecom Business Supporting Based On Cloud Computing Platform

Business Support System for Telecommunication holds up operation intergration of Telecommunication, information resources sharing, is the key to success of every major telecommunication service provider in current competitive circumstance. Based on grid computing, utility computing and parallel computing, cloud computing is a new distributed computing technology. It provides platform for large scale resources deployment and management. In these processes, whether providing right resources to virtual machines very vital. Cloud computing has the characteristics of a wide variety of resources and heterogeneity between resources. Existing resources provision policies do not fully take heterogeneity between resources into account, so these policies are not fully fit for complex heterogeneous environment.For characteristics of Business Support System for Telecommunication, this paper, based on these existing resources provision policies, presents a EARC (Economy-Availability-Reliability-Collaboration)model based resources provision policy. This model takes economy , availability, reliability, collaboration of resource into accout.the calcutation model of economy factor constituted by the resource cost of hardware and energy costs,reduce the cost of cloud providers's operations. the calcutation model of availability factor constituted by utilization of existing resources and the remaining available computing power,the former improves load balancing of physical resources,the latter improves localization feature of resource scalability. Reliability factor is reflected by resource reliability status through resource predictor, evey reliable or unreliable resource, which meets the requirements of basic calculation task will be included in the resource pool of candidates,further improving overall resource utilization.For collaboration factor ,difference in performance for different physical resources to perform different types of tasks is taken into accout. Through combination of historical information to measure the collaboration index, this model could improve the performance of resources when tastk are executed. This model can adjust the overall performance through imposing different weight on resources.Finally,by simulation experiments,the effectiveness of EARC resource provisioning policy has been verified.