命名数据网络中的一种主动拥塞控制策略

Active congestion control strategy in named data networking

为解决命名数据网络(Named Data Networking,NDN)中的拥塞控制问题,在加权公平排队(Weighted Fair Queuing,WFQ)算法的基础上,提出了基于即时调整兴趣包发送速率的名字加权公平排队(Name Weighted Fair Queuing,NWFQ)算法。当网络拥塞时,利用基于令牌桶算法的速率限制机制对超速流进行降速惩罚,同时更新兴趣包中的拥塞信息域并将其逐跳反馈给下游路由器;利用数据包将拥塞信息反馈给请求端,请求端据此调整兴趣包的发送速率,从而解决网络拥塞。基于ndn SIM的仿真结果表明,该算法能有效提升瓶颈链路利用率并接近95%,同时保证较低的丢包率和平均流完成时间。

This paper proposes a Name Weighted Fair Queuing(NWFQ)algorithm on the basis of the Weighted Fair Queuing(WFQ)algorithm,NWFQ timely changing the sending rate of interest packets to solve the congestion control problem in Named Data Networking(NDN).Firstly,when congestion occurs,the speed limit mechanism based on token bucket algorithm is used to reduce the forwarding rate of the overspeed flow.At the same time,the congestion information field in the interest package is updated and then hop by hop sended to the downstream router;finally,data packets encapsulate the congestion information and return to the consumer,by this way,consumer can adjust the interest sending rate and the data rate accordingly to mitigate network congestion.Simulation results based on ndnSIM show that:NWFQ can effectively improve the bottleneck link utilization and close to 95%,while maintaining a low packet loss rate and average flow completion time.