节点文献

信息中心网络中基于流量负载控制的缓存策略

A Caching Strategy Based on Control of Traffic Load in the Information-Centric Networking

  • 推荐 CAJ下载
  • PDF下载
  • 不支持迅雷等下载工具,请取消加速工具后下载。

【作者】 杨奇峰邓浩江王玲芳

【Author】 YANG Qifeng;DENG Haojiang;WANG Lingfang;Network New Media Engineering Research Center,Institute of Acoustics,Chinese Academy of Sciences;University of Chinese Academy of Sciences;

【机构】 中国科学院声学研究所国家网络新媒体工程技术研究中心中国科学院大学

【摘要】 信息中心网络通过高速网内缓存可以有效降低用户的访问时延,降低服务器和链路负载,提升网络性能。然而,网络中部分节点负载过高导致节点内缓存替换率升高,使得个别节点乃至整体网络性能下降。本文提出一种基于流量负载控制的缓存策略,根据节点接收队列的流量状态,动态控制节点缓存的替换速率,并将由于负载控制而无法缓存的内容交给下游低负载节点缓存。实验结果显示,我们的策略可以有效提升高流量负载节点的缓存命中率,同时提高了低负载节点的缓存利用率,均衡了网络中的缓存负载状态,提升了网络的整体性能。

【Abstract】 Information-centric networking can effectively reduce user delay, server load and link consumption through high-speed in-network cache. However, due to the overload of caching nodes, the replacement rate in these nodes can be too high to guarantee the performance of the nodes and even affects the overall network performance. This paper proposes a caching strategy based on control of traffic load. According to the status of the receiving queue of a node, the replacement rate of the node is dynamically controlled. The content that cannot be cached because of the limited replacement rate is delivered to the downstream idle nodes. According to the result of our experiment, our strategy can help high-load nodes obtain a higher cache hit rate, while improving the utilization ratio of low-load nodes. The caching efficiency in the network is balanced and the overall network performance is improved by implementing our strategy.

【基金】 中国科学院战略重点研究计划:SEANET技术的标准化研究与系统开发(编号:XDC02070100)
  • 【文献出处】 网络新媒体技术 ,Network New Media Technology , 编辑部邮箱 ,2021年05期
  • 【分类号】TP393.06
  • 【下载频次】50
节点文献中: 

本文链接的文献网络图示:

本文的引文网络