ICN中支持流媒体业务的边缘缓存与循迹路由混合机制  

Hybrid Mechanism Based on Edge Caching and Tracing Route Supporting Streaming Media Services in Information Centric Networks

作　　者：田铭[1] 杨森[1] 兰巨龙[1] 

机构地区：[1]国家数字交换系统工程技术研究中心,河南郑州450002

出　　处：《信息工程大学学报》2016年第6期735-742,共8页Journal of Information Engineering University

基　　金：国家自然科学基金资助项目(61309019;61372121);国家973计划资助项目(2012CB315901;2013CB329104);国家863计划资助项目(2015AA016102)

摘　　要：针对流媒体业务特点,提出一种基于边缘存储与循迹路由的混合机制(HECT),将流媒体业务数据chunk以渐进式概率缓存推送至网络边缘进行存储,并利用循迹路由对后续内容请求进行响应,减少核心网络的负载压力;利用流媒体业务数据chunk之间的相关性,主动并行发送后续数据chunk的预测请求,执行临时缓存。仿真实验和结果表明,HECT算法具有较小的平均响应时延,减小了服务器响应率和平均路由跳数,最后给出了算法对内容请求的集中分布性和内容相关性的适应性讨论。According to the characteristics of the streaming media services, the paper puts forward a hybrid mechanism based on Edge Caching and Tracking routing （HECT）. The streaming data chunks are pushed to the edges of the network with progressive probability, then the subsequent content requests can be responded by tracking routing to reduce the load pressure of the core network. To retrieve and cache content in advance, parallel requests of relevant data units are sent initiatively based on the correlation characteristics of content request sequences. Simulation and experimental results show that the HECT algorithm has a smaller average response delay, reduces the server response rate and the average route hop. The adaptability of the algorithm to intensive distribution and correlation is discussed in the end.

关 键 词：信息中心网络 流媒体业务 概率缓存 循迹路由 相关性预测 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]