基于双边匹配方法的交换机迁移策略研究  

Research on Switch Migration Strategy Based on Two-Sided Matching Method

在线阅读下载全文

作  者:黄陈 琚贇[1] 胡州明 祝文军 HUANG Chen;JU Yun;HU Zhouming;ZHU Wenjun(School of Control and Computer Engineering,North China Electric Power University,Beijing 102206;Aostar Information Technologies Co.,Ltd.,Chengdu 610041;Beijing Zhongdian Puhua Information Technology Co.,Ltd.,Beijing 100192)

机构地区:[1]华北电力大学控制与计算机工程学院,北京102206 [2]四川中电启明星信息技术有限公司,成都610041 [3]北京中电普华信息技术有限公司,北京100192

出  处:《计算机与数字工程》2025年第2期455-459,共5页Computer & Digital Engineering

基  金:国家重点研发计划项目(编号:2020YFB0905900)资助。

摘  要:软件定义网络中的多控制器部署提高了控制平面的可扩展性和可靠性,但由于流量的突变性会导致多控制器间的负载不平衡,通过交换机迁移策略可以有效解决负载不平衡问题,但现有方法还存在迁移效率低,迁移次数过多导致迁移成本增加等问题。针对这一问题,提出了一种基于双边匹配的交换机迁移策略SMTM(Switch Migration Strategy based onTwo-Sided Matching)。通过在策略中添加负载预测组件,并结合当前负载和下一时刻负载判断控制器过载情况,依据双边匹配的方法挑选出最佳的待迁移交换机和目标控制器。实验结果表明,相比于现有方法,所提策略能有效降低迁移次数,提高控制器性能,实现控制器间的负载均衡。Multi-controller deployments in software defined networks improve the scalability and reliability of the control plane,but the sudden variability of traffic can lead to load imbalance among multiple controllers.The load imbalance problem can be effectively solved by a switch migration strategy,but existing methods still have problems such as low migration efficiency and in⁃creased migration costs due to too many migrations.To address this problem,a Switch Migration Strategy based on Two-Sided Matching(SMTM)is proposed.By adding a load prediction component to the strategy and combining the current load and the next load to determine the controller overload,the best switch and target controller to be migrated are selected based on the two-sided matching method.Experimental results show that the proposed strategy can effectively reduce the number of migrations,improve controller performance and achieve load balancing between controllers compared to existing methods.

关 键 词:软件定义网络 多控制器 交换机迁移 双边匹配 负载均衡 

分 类 号:TP393.0[自动化与计算机技术—计算机应用技术]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象