内容中心网络中基于区域集中化控制的协作缓存路由机制

Cooperative caching and routing scheme based on local centralized control in content centric networking

针对在内容中心网络(content centric networking,CCN)中如何合理放置与高效利用应答数据的问题,将集中化控制的思想引入到内容缓存与查找中,提出一种协作缓存路由机制。在缓存决策时,通过兴趣包和数据包携带标签的方式,确定沿途最大缓存收益区域;在最大缓存收益区域内,结合内容全局活跃度和节点可用缓存空间,选择内容最佳放置位置。在路由查找时,将区域内容放置与路由转发相结合,增大缓存资源可用性。仿真结果表明,与经典算法相比,该机制以少量额外的开销提高了缓存命中率和跳数减少率,改善了缓存负载分布均衡性,提升了CCN网络缓存和传输效率。

Aiming at the reasonable placement and efficient use of response data in the content centric network （CCN） , this paper proposed a cooperative caching and routing scheme based on local centralized control （CRLCC）. Along the content de- livery path, it determined the subsystem which had the biggest caching gain by matching labels in data packets. In subsystems, the proposed scheme selected the optimal placement according to available cache space of nodes and content activity. When a content was requested, this scheme looked for the requested content by combining local content placement with route forwarding. Simulation results show that this scheme can achieve higher cache hit ratio and hop reduction ratio, and make the cache distribution among nodes more uniform.