基于SDN的数据中心网络流量调度机制的设计与实现  

Design and Implementation of Flow Scheduling Mechanism Based on SDN for Data Center Network

作　　者：王文涛[1] 郑芳[1] 王玲霞[1] 穆晓峰[1] 

机构地区：[1]中南民族大学计算机科学学院,武汉430074

出　　处：《中南民族大学学报（自然科学版）》2016年第3期135-140,共6页Journal of South-Central University for Nationalities：Natural Science Edition

基　　金：国家民委教改基金资助项目(15013);中南民族大学研究生创新基金资助项目(2016sycxjj199)

摘　　要：针对SA算法中未考虑当前网络链路带宽资源引起的流冲突问题以及GFF算法中未考虑流带宽需求变化引起带宽资源分配不合理问题,提出了基于模拟退火遗传算法的按需自适应(SAGA-AO)流量调度机制.该机制首先依据流带宽需求变化筛选出网络中需要调度的流,然后利用模拟退火遗传算法(SAGA)根据当前链路带宽资源状况对需要调度的流进行全局调度路径搜索.仿真结果表明:SAGA-AO算法在大多数通信模型下平均对分带宽高于SA和GFF算法.SA algorithm has the flow conflict problem because it does not consider the current network link bandwidth resources, and GFF algorithm easily lead to the irrational distribution of bandwidth resources because of not considering the change of flow demand. To solve the two problems, we proposed an adaptive on-demand flow scheduling mechanism. First, the mechanism filtered the flows needed to be scheduled based on the changing of demand. Second, it globally searched scheduling path for these flows using the simulated annealing genetic algorithm （ SAGA） based on the available link bandwidth resources. The simulated results showed that the mechanism we proposed outperform GFF and SA in the majority of communication patterns.

关 键 词：数据中心网络 流量调度 按需自适应 模拟退火遗传算法 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]