Storage and computing resource enabled joint virtual resource allocation with Qo S guarantee in mobile networks  

作　　者：Xiaodong XU Jiaxiang LIU Wenwan CHEN Yanzhao HOU Xiaofeng TAO 

机构地区：[1]National Engineering Laboratory for Mobile Network Technologies,Beijing University of Posts and Telecommunications, Beijing 100876, China

出　　处：《Science China(Information Sciences)》2017年第4期45-55,共11页中国科学（信息科学）（英文版）

基　　金：supported by International Collaboration Project (Grant No. 2015DFT10160);National Natural Science Foundation of China (Grant Nos. 61471068, 61421061, 61325006);National High-Tech R&D Program of China (863) (Grant No. 2014AA01A701);National Major Project (Grant No. 2016ZX03001009-003);Beijing Training Project for the Leading Talents in S&T (Grant No. Z141101001514026);111 Project of China (Grant No. B16006)

摘　　要：Virtualization is the trend for the future mobile networks. With the advantage of virtualization,we can abstract the physical mobile network into the virtual network function（VNF） and design the network without the details. In this paper, we focus on the virtualization of the physical resources so that the resource allocation scheme considers not only the time-varying characteristic of wireless channels but also the amount of storage and computing resources. Virtual resources are composed of radio, storage and computing resources based on the virtualization technology. Since the cloud radio access network（C-RAN） is a successful paradigm to introduce computing resources into mobile networks, we investigate the virtual resource allocation scheme in the C-RAN architecture. With the content caching technology, we introduce the storage resources into joint resource allocation scheme further. In order to evaluate the performance of proposed scheme, we choose the effective capacity as the metric to include the influence of service latency. The purpose of the optimization problem is maximizing the system effective capacity with constraints of radio, storage and computing resources.It is simplified and converted into a convex problem solved by the subgradient method. Simulation results are provided to demonstrate performance gain of the effective capacity based joint resource allocation scheme.Virtualization is the trend for the future mobile networks. With the advantage of virtualization,we can abstract the physical mobile network into the virtual network function（VNF） and design the network without the details. In this paper, we focus on the virtualization of the physical resources so that the resource allocation scheme considers not only the time-varying characteristic of wireless channels but also the amount of storage and computing resources. Virtual resources are composed of radio, storage and computing resources based on the virtualization technology. Since the cloud radio access network（C-RAN） is a successful paradigm to introduce computing resources into mobile networks, we investigate the virtual resource allocation scheme in the C-RAN architecture. With the content caching technology, we introduce the storage resources into joint resource allocation scheme further. In order to evaluate the performance of proposed scheme, we choose the effective capacity as the metric to include the influence of service latency. The purpose of the optimization problem is maximizing the system effective capacity with constraints of radio, storage and computing resources.It is simplified and converted into a convex problem solved by the subgradient method. Simulation results are provided to demonstrate performance gain of the effective capacity based joint resource allocation scheme.

关 键 词：VIRTUALIZATION resource allocation CACHE computing resource effective capacity C-RAN 

分 类 号：TN929.5[电子电信—通信与信息系统]