Optimizing TTL Caches under Heavy-Tailed Demands
Optimizing TTL Caches under Heavy-Tailed Demands
复制标题
DOI：
10.1145/2896377.2901459
复制
发表时间：
2016-06
期刊：
影响因子：
--
通讯作者：
Andrés Ferragut;Ismael Rodríguez;F. Paganini
中科院分区：
文献类型：
--
作者：
Andrés Ferragut;Ismael Rodríguez;F. Paganini
在本文中，我们分析了以一般到达分布和不同受欢迎程度接收文件请求的高速缓存系统的命中性能。我们考虑基于计时器的（TTL）策略，并具有优化的差异化计时器。最佳策略证明与到达分布的危险率函数的单调性有关。尤其是降低危险率，计时器政策的表现优于缓存最受欢迎内容的静态政策。在帕累托分布的请求时间和文件受欢迎程度的ZIPF分布的情况下，我们为最佳策略提供明确的解决方案，包括大量文件的限制的紧凑流体表征。我们通过模拟与经典策略（例如最少使用的策略）进行比较，并讨论其性能。最后，我们将优化框架的扩展分析到缓存线网络。
In this paper we analyze the hit performance of cache systems that receive file requests with general arrival distributions and different popularities. We consider timer-based (TTL) policies, with differentiated timers over which we optimize. The optimal policy is shown to be related to the monotonicity of the hazard rate function of the inter-arrival distribution. In particular for decreasing hazard rates, timer policies outperform the static policy of caching the most popular contents. We provide explicit solutions for the optimal policy in the case of Pareto-distributed inter-request times and a Zipf distribution of file popularities, including a compact fluid characterization in the limit of a large number of files. We compare it through simulation with classical policies, such as least-recently-used and discuss its performance. Finally, we analyze extensions of the optimization framework to a line network of caches.