| Internet routers require buffers to hold packets during times of congestion. The buffers need to be fast, and so ideally they should be small enough to use fast memory technologies such as SRAM or all-optical buffering. Unfortunately, a widely used rule-of-thumb says we need a bandwidth-delay product of buffering at each router so as not to lose link utilization. This can be prohibitively large.; In a recent paper, Appenzeller et al. challenged this rule-of-thumb and showed that for a backbone network the buffer size can be divided by sqrt( N) without sacrificing throughput, where N the number of flows sharing the bottleneck. In this dissertation, we explore how buffers in the backbone can be significantly reduced even more, to as little as a few dozen packets, if we are willing to sacrifice a small amount of link capacity. We argue that if the TCP sources are not overly bursty, then 20-50 packet buffers are sufficient for high throughput. Specifically, we argue that O(log W) buffers are sufficient, where W is the congestion window size of each flow. We support our claim with analysis, a variety of simulations, and some experiments in real networks.; The change we need to make to TCP is minimal---each sender just needs to pace packet injections from its window. Moreover, there is some evidence that such small buffers are sufficient even if we do not modify the TCP sources so long as the access network is much slower than the backbone, which is true today and likely to remain true in the future.; We conclude that buffers can be made small enough for all-optical routers with small integrated optical buffers. |
