基于路径代价和节点代价的信息中心网络缓存策略

Path Cost and Node Cost Based Caching Strategy for Information-centric Network

网络化缓存是ICN(信息中心网络)架构的重要特征之一,对改善网络性能起着重要作用,如何合理利用有限的缓存空间,在优化缓存部署时实现缓存开销最小是一个值得关注的问题.考虑到缓存开销不仅包含节点访问缓存节点时的路径代价,还应包含被新缓存内容所替换的旧内容的替换代价,提出了基于路径访问代价和节点替换代价的缓存策略(Path and Node Cost based Caching Strategy,简称PNCCS).该文首先建立了一个理论模型来分析缓存部署对路径访问代价和节点替换代价的影响.基于这个模型,缓存部署问题可以形式化地描述成一个最优化问题,并采用离散粒子群优化算法进行求解,最优解代表着一种优化的缓存部署方案.实验结果表明,PNCCS策略与CEE、Prob和LCD相比,在降低网络运行开销的同时提升了用户体验质量两方面的性能指标.

In-network caching,a typical feature of information centric networking（ ICN） architecture,has played an important role in improving the network performance. One of the important issues with in-network caching is coordinating cache placement to use cache resource maximization and achieve access cost minimization. Considering the cache cost not only contains the path cost raised by hop count between the require node and cache node but also the replacement cost,in this paper,a Path and Node Cost based Caching Strategy（ PNCCS） is proposed. A theoretical model is introduced to analyze the path cost and the replacement cost of caching content in the network,in this model,the cache placement problem is formulated as an optimization problem. A discrete particle swarm optimization is proposed to solve the problem,the optimal solution corresponds to an optimal cache placement. The simulation experiments demonstrate that the proposed PNCCS,compared with CEE,Prob and LCD,has significantly improved the network performance in two aspects,i. e.,the operating cost and the quality of user experience.