摘要
现有的资源分配算法研究主要面向数据中心和云计算环境展开,未能充分考虑基站处理资源的多样性(CPU、内存、网络带宽、FPGA、DSP)和业务种类的多样性,不能直接应用于集中式接入网架构中。针对该问题,提出了基于业务类型的资源分配算法,首先采用Fisher分割方法,根据用户业务类型对不同类型计算资源的需求,对业务进行分类,然后利用资源分配均衡策略分配基站处理资源。仿真结果表明,该算法有效地减少了开启物理服务器的个数并提高了物理服务器的资源利用率,达到了绿色节能的目的。
Current research on resource allocation algorithms mainly focuses on data centers, and cloud computing environments. These resource allocation algorithms did not consider the diversity of processing resources（CPU, memory, network bandwidth, FPGA, DSP） and the diversity of service types in the centralized base station, resulting in the low processing resource utilization. In order to solve this problem, a service oriented base station processing resource allocation algorithm was proposed. Firstly, the Fisher partition method was used to classify the base station＇s processing resource requirements according to the user＇s service request. Then, the resource allocation balancing strategy was used to allocate the base station processing resources. Experimental results show that the algorithm is effective to reduce the number of physical servers and improves the resource utilization of the servers, realizing energy conservation and communications.
作者
张新苹
王园园
田霖
郝树良
ZHANG Xinping;WANG Yuanyuan;TIAN Lin;HAO Shuliang(Chongqing University of Posts and Telecommunications,Chongqing 400065,China;Institute of Computing Technology,Chinese Academy of Sciences,Beijing 100190,China)
出处
《电信科学》
2018年第8期109-118,共10页
Telecommunications Science
基金
国家自然科学基金资助项目(No.61431001)~~
关键词
集中式接入网
基站处理资源分配
用户业务类型
Fisher分割方法
资源分配均衡策略
centralized radio access network
base station processing resource allocation
user service type
Fisherpartition method
resource allocation balancing strategy
