面向5G的分布式移动云计算协同架构与管理机制  

Cooperative Management Architecture and Mechanism of 55 G-Oriented Distributed Mobile Cloud Computing G-Oriented

作　　者：李宏佳[1] 陈鑫[1] 周旭[1] 

机构地区：[1]中国科学院,北京100093

出　　处：《中兴通讯技术》2015年第2期14-19,共6页ZTE Technology Journal

基　　金：国家自然科学基金(61302108)

摘　　要：提出了一种分布式移动云计算协同架构以及端到端终端计算卸载服务质量保证机制。该架构将云计算资源嵌入移动通信网络基站等网元,形成近端低时延云计算能力;引入分布式移动云计算协同控制器,以及基于其的终端、移动网络与分布式云服务器协同的感知与计算卸载决策机制。该架构可以降低终端在计算卸载中的网络交互信令开销,以及终端驻留决策进程的CPU占用率与能耗;可以保证移动云计算的端到端服务质量;可以规避传统移动云计算给广域网(WAN)带来的巨大负载量与能耗。In this paper, we propose a new distributed mobile computing over 5G networks and a cooperative management framework for guaranteeing end-to-end quality of service of terminal computation offloading. Specifical y, we design cloudlets that have only a smal fraction of the computational ability of the cloud computing center. These cloudlets are associated with eNodeBs to reduce transit delay through the WAN. A cooperative control er is introduced to determine the computation offloading policy according to the key factors measured from smart terminals, the RAN and cloudlets. Compared with the classical mobile cloud computing, the proposed structure and cooperative management framework reduces overhead, CPU occupancy rate, and energy consumption of smart terminals before and during the computation offloading. It also guarantees end-to-end QoS of distributed mobile cloud computing and avoids huge overhead and energy consumption.

关 键 词：移动云计算 服务质量 智能终端 协同 时延 能耗 

分 类 号：TN929.5[电子电信—通信与信息系统] TP393[电子电信—信息与通信工程]