结合内容流行和主观偏好的节点缓存策略

Node Cache Strategy Combining Content Popularity and Subjective Preference

近年来,随着内容分发要求的不断提高,互联网逐渐从以主机为核心演变为面向内容模式.作为内容中心网络的典型代表,命名数据网络受到了越来越多关注.针对命名数据网络中的高效缓存内容问题,提出一种节点主观视角下结合内容流行和节点偏好的缓存策略.具体为每个节点通过周期性地统计自己收到和发出的Interest包信息来计算每类内容在该节点主观视角下的流行度和该节点对不同内容类型的偏好度,以得到每类内容的基本优先级,进一步在收到Data包时,结合内容大小评估值计算每个内容的缓存优先级,进而执行缓存策略.该缓存策略计算方法简单易行,无需节点间额外通信交换信息.仿真结果表明,该缓存策略比主流策略具有更高的缓存命中率、更低的缓存替换率和更近的请求距离.

In recent years,with the gradual increment of content distribution requirements,Internet evolves from host-centric to content-oriented.As a typical representative of information-centric networking,Named Data Networking （NDN） is attracting more and more attention.To solve the problem of caching content efficiently in NDN,a cache strategy from subjective viewpoint combining content popularity and node preference is proposed.Each node calculates the popularity degree and preference degree of each content type from own subjective viewpoint through counting the received and sent Interest packets periodically respectively,and then gets the basic priority of each content type.When receiving a Data packets,each node combines the basic priority of content type and content size evaluation to calculates the content cache priority and execute corresponding caching strategy.The strategy is simple and easy to implement,and does not need the extra communication among nodes.The simulation results show that it has higher cache hit rate,lower cache replacement rate,and nearer request distance than the popular cache strategies.