论文部分内容阅读
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.
Network caching is one of the most important features of ICN (Information Center Network) architecture. It plays an important role in improving network performance. How to use limited cache space reasonably and minimize cache overhead when optimizing cache deployment is a concern. Considering that the cache overhead not only includes the path cost when the node accesses the cache node but also the replacement cost of the old content replaced by the new cache content, a cache policy based on the path access cost and the node replacement cost (Path and Node Cost based Caching Strategy (referred to as PNCCS) .First, this paper establishes a theoretical model to analyze the impact of cache deployment on path access cost and node replacement cost.Based on this model, the cache deployment problem can be formally described as an optimization problem, Discrete particle swarm optimization algorithm to solve the optimal solution represents an optimized cache deployment scheme.The experimental results show that compared with CEE, Prob and LCD, PNCCS strategy improves the quality of user experience while reducing network overhead The performance index.