信息中心网络中一种基于内容热度的分区缓存替换方法  

A partition cache replacement method based on content popularity in information-centric networking

作　　者：李远航 王劲林[1,2] 韩锐 LI Yuanhang;WANG Jinlin;HAN Rui(Insitute of Acoustics,Chinese Academy of Sciences,Beijing 100190,China;University of Chinese Academy of Sciences,Beijing 100049,China)

机构地区：[1]中国科学院声学研究所,北京100190 [2]中国科学院大学,北京100049

出　　处：《电子设计工程》2023年第6期133-138,143,共7页Electronic Design Engineering

基　　金：中国科学院战略性先导科技专项课题(XDC02070100)。

摘　　要：针对传统的LRU(Least Recently Used)替换策略在信息中心网络的缓存场景中存在难以持久缓存热门内容、缓存性能提升有限的问题,提出了一种基于内容热度的分区缓存替换方法。该方法将缓存空间分成热门区、中间区和冷门区,根据新内容的热度将其插入到对应分区,并在发生缓存替换时选择冷门区的低热度内容进行替换。通过与LRU、先进先出(First Input First Output,FIFO)、随机替换(Random Replacement,RR)方法进行对比仿真实验,发现该文所提出的方法能够降低热门内容被替换的概率,并显著提高了缓存命中率,降低了内容获取时延。Aiming at the problems that the traditional LRU(Least Recently Used) replacement strategy is difficult to cache popular content for a long time and the cache performance is limitedin the cache scenario of information-centric networks,a partition cache replacement method based on content popularity is proposed in this paper. In the proposed method,the cache space is divided into popular,middle and unpopular areas. New content is inserted into the corresponding partition according to the popularity of the new content,and the low-popularity content in the unpopular area is selected for replacement when the cache replacement occurs. Compared with LRU,FIFO(First Input First Output) and RR(Random Replacement) methods,the simulation results show that the proposed method can reduce the probability of popular content being replaced and significantly improve the cache hit ratio. The content acquisition delay is also reduced.

关 键 词：网内缓存 信息中心网络 缓存替换 LRU 

分 类 号：TP393.0[自动化与计算机技术—计算机应用技术]