摘要
针对目前用于IP路由查找的地址缓存技术和前缀缓存技术的局限性,分析了骨干网路由表前缀重叠特征,提出了一种基于阈值的IP路由缓存方法,该方法结合了地址缓存和前缀缓存技术,无需进行前缀扩展,克服了地址缓存技术缓存空间要求过大、前缀缓存技术无法缓存内部前缀节点的问题,在缓存空间、缓存命中率、缓存公平性以及路由增量更新方面具有优势;仿真实验表明对于路由条目超过260000的路由表,缓存空间大小为30000,选择阈值K=4时97%以上的节点可实现1:1缓存,其余节点采用地址缓存,缓存失效率小于0.02,可以用小的缓存空间实现高速线速转发.
The current IP address cache or IP prefix cache technologies have limitations.We analyse the overlapping relationships among prefixes in the global routing tables and propose a threshold-based routing cache method which combines IP address cache and prefix cache technologies without the need of prefix expansion.The threshold value K is selected according to the prefix overlapping features of global routing tables,which overcomes the shortcomings of address caching and prefix caching technologies.Comparison and simulation show that our scheme has better performance over other schemes in cache size,cache hit ratio,fairness among prefixes and incremental prefix updates.For a global routing table with more than 260 000 entries and cache size of 30 000,above 97% prefix nodes could be 1：1 cached by prefix cache and the other prefix nodes could be cached by address cache with threshold K=4.Computation results show that high-speed forwording could be fulfilled with small cache size.
出处
《计算机研究与发展》
EI
CSCD
北大核心
2012年第4期710-716,共7页
Journal of Computer Research and Development
基金
国家"八六三"高技术研究发展计划基金项目(2005AA121410)
电子基金项目(04C26224210719)
关键词
路由查找
路由缓存
地址缓存
前缀缓存
前缀扩展
阈值
IP address lookup
routing caching
address cache
prefix cache
prefix expansion
threshold
