北京邮电大学学报

  • EI核心期刊

北京邮电大学学报 ›› 2021, Vol. 44 ›› Issue (2): 68-74.doi: 10.13190/j.jbupt.2020-086

• 未来网络体系架构和关键技术专题 • 上一篇    下一篇

面向新生内容需求的缓存放置与替换联合算法

单思洋, 冯春燕, 朱光宇, 张天魁   

  1. 北京邮电大学 信息与通信工程学院, 北京 100876
  • 收稿日期:2020-07-07 出版日期:2021-04-28 发布日期:2021-04-28
  • 通讯作者: 张天魁(1980-),男,教授,E-mail:zhangtiankui@bupt.edu.cn. E-mail:zhangtiankui@bupt.edu.cn
  • 作者简介:单思洋(1988-),男,博士生.
  • 基金资助:
    国家自然科学基金项目(61971060,61502046)

A Joint Cache Placement and Replacement Algorithm for Cache Requirements of New Contents in CCN

SHAN Si-yang, FENG Chun-yan, ZHU Guang-yu, ZHANG Tian-kui   

  1. School of Information and Communication Engineering, Beijing University of Posts and Telecommunications, Beijing 100876, China
  • Received:2020-07-07 Online:2021-04-28 Published:2021-04-28

摘要: 在内容中心网络中,全局缓存放置算法无法对新生内容实时缓存放置进行优化,对此,提出一种缓存放置与替换的联合优化算法,可减小缓存节点业务负载和用户内容获取的时延.定义了整网缓存收益函数,构建了最大化整网缓存收益的最优化问题,以实现新生内容的缓存放置与已存内容的缓存替换.为了求解所提的优化问题,将所提优化问题分解为缓存放置子问题和缓存替换子问题,提出了一种全局缓存放置优化问题的次优解,缓存放置算法的性能下限为(1-1/e)倍的最优解.在缓存放置之后考虑多点协同的缓存替换算法,最小化由于缓存替换产生的缓存损失,最后通过迭代实现缓存放置与替换的联合优化.仿真结果表明,所提联合优化算法可以提升整网缓存收益,在缓存节点负载、内容获取平均跳数和全网缓存命中率方面均优于传统方法.

关键词: 内容中心网络, 缓存放置算法, 缓存替换算法

Abstract: In content centric networking (CCN), in order to solve the problem that the existing global cache placement algorithm cannot optimize the cache placement for new content in real time, the global cache placement and replacement algorithm for new content cache requirements is studied, reducing the cache node load and content acquisition delay for users. The network-wide cache gain function is defined, and the optimization problem of maximizing the entire network cache gain is formulated to realize the cache placement of new contents and the cache replacement of existing contents. The optimization problem is decomposed into a cache placement sub-problem and a cache replacement sub-problem. A suboptimal solution to the global cache placement optimization problem is proposed. A collaborative cache replacement algorithm to minimize the loss of cache gain due to cache replacement is proposed. Finally, the joint optimization of cache placement and replacement is realized through iterations. Simulation shows that the proposed algorithm is superior to the traditional methods in terms of cache node load, user hop count and cache hit ratio.

Key words: content centric networking, caching placement algorithm, caching replacement algorithm

中图分类号: