Journal of Frontiers of Computer Science and Technology ›› 2015, Vol. 9 ›› Issue (10): 1229-1237.DOI: 10.3778/j.issn.1673-9418.1412067

Previous Articles     Next Articles

Technology of Deploying In-Network Cache Based on Content Popularity

JIN Jing+   

  1. Department of Computer Science and Technology, Tsinghua University, Beijing 100084, China
  • Online:2015-10-01 Published:2015-09-29

基于内容流行度的网内缓存部署技术

金晶+   

  1. 清华大学 计算机科学与技术系,北京 100084

Abstract:  In-network cache is a basic service of information-centric networking. In order to deploy in-network cache efficiently, this paper completes mainly two works: (1) Study the characteristics of the domestic video content distribution. It shows a highly concentrated characteristic with a stable trend in both content space and geographical space. (2) Propose a technology of deploying in-network cache based on content popularity in order to reduce the cache redundancy, improve cache difference rate and optimize the average access cost with limited cache redundancy. The experimental results show that the proposed technology has a lower average access cost than similar whole network cooperative caching technology. It also slightly improves cache difference rate and cache hit rate.

Key words: in-network cache, caching strategy, caching coordination, regional popularity

摘要: 网内缓存是信息中心网络的基础服务之一,为了实现高效率的网内缓存部署,主要进行了两方面的工作:一是对国内视频内容的流行分布特征进行了研究,发现无论是在内容空间上还是在地域分布上,国内视频内容的流行分布呈现出高度集中的特征,并且趋势比较稳定。二是提出了一种基于内容流行度的网内缓存部署技术,用于降低网内缓存冗余度,提高缓存差异率,同时优化有限缓存冗余下的平均接入代价。实验结果表明,该技术比同类型的全网协同缓存技术具有更低的平均接入代价,并在缓存差异率和缓存命中率方面有小幅提升。

关键词: 网内缓存, 缓存策略, 协同缓存, 区域流行度