دانلود رایگان مقاله ذخیره در حافظه پنهان FIB روتر عملگرا

Abstract

Several recent studies have shown that router FIB caching offers excellent hit rates with cache sizes that are an order of magnitude smaller than the original forwarding table. However, hit rate alone is not sufficient – other performance metrics such as memory accesses, robustness to cache attacks, queuing delays from cache misses, etc., should be considered before declaring FIB caching viable. In this paper we tackle several pragmatic questions about FIB caching. We characterize cache performance in terms of memory accesses and delay due to cache misses. We study cache robustness to pollution attacks and show that in order to evict the most popular prefixes an attacker must sustain packet rates higher than the link capacity. We show that caching was robust even during a recent flare of NTP attacks. We carry out a longitudinal study of cache hit rates over four years and show the hit rate is unchanged over that duration. We characterize cache misses to determine which services are impacted by FIB caching. We conclude that FIB caching is viable by several metrics, not just impressive hit rates.

9. Conclusions

In this paper we take a look at some practical considerations in the implementation of FIB caching. We extend previous work in significant ways by looking at practical issues, such as charac- terization of cache misses, queuing issues (buffer occupancy and delay), memory bandwidth requirements and robustness against 652 cache pollution attacks. We used traces collected at links from our regional ISP to a Tier-1 ISP for our analysis. Our design uses a cacheable FIB to avoid the cache hiding problem, and we presented an algorithm to convert a regular FIB to a cacheable FIB. Our work has some limitations. First, we only look at packet traces from a single regional ISP. We therefore cannot evaluate cache performance at core routers, where traffic may be more di- verse causing hit rates to drop. While we do not have access to data from core routers to answer this question (we need a packet 661 trace and a simultaneous snapshot of the FIB at a core router), the tools and methodology we developed are applicable to a core en- vironment and we plan to repeat the study once we have an ap- propriate dataset. Are there trends that may invalidate the benefits of FIB caching? On the contrary, recent trends such as traffic concentra- tion to major social networks, search engines that reside in data- centers [14] and CDNs make caching even more likely to provide benefits. In these environments, traffic becomes more focused and likely to hit a smaller set of prefixes, resulting into a more stable working set.