摘要： 内容中心网络(CCN)作为一种主要的未来网络架构,以命名的内容作为网内的主要元素之一,在网络研究中受到广泛关注。针对已有的CCN 缓存方案内容副本替换严重的问题,提出一种内容热门度与节点中介中心度约束的缓存机制PopBetw。在基于节点中心度的基础上,从内容本身的属性热门度出发,避免非热门内容的不必要缓存,降低每个节点的缓存负荷,提高网络缓存性能。仿真结果表明,通过评估缓存大小和内容热门度对缓存性
能的影响,PopBetw 缓存策略可取得比LCE,LCPro 和EgoBetw 方案更高的缓存命中率和更小的路径延展度,明显降低网内缓存替换数量,有效减少网内节点中介中心性较大节点群的缓存替换数,达到整体性能最优化。
关键词:
内容中心网络,
缓存,
内容热门度,
中介中心度,
缓存替换,
缓存冗余
Abstract: Content-centric Network(CCN) based on named data which is one of the vital elements in network,as a
main architecture for the future Internet,attracts considerable attention from the research community recent years. A new caching scheme called PopBetw based on content popularity and node’s betweenness is proposed to solve the problem of heavy replacement of content copies brought by existing cache policies in CCN:combining content popularity which is one of content attributes with node’s betweenness avoids caching unpopular contents,reduces caching load in each node and improves in-network caching performance. Simulation experiment demonstrates that the proposed PopBetw caching strategy achieves larger cache hit rate and lower path stretch than that of LCE scheme,LCPro and EgoBetw scheme by evaluating the impact of cache size and content popularity on the caching performance,it also shows that PopBetw apparently lower the number of in-network cache replacement especially in nodes with relatively large value of betweenness to achieve better overall performance.
Key words:
Content-centric Network(CCN),
caching,
content popularity,
betweenness,
cache replacement,
cache redundancy
中图分类号: