Skip to Main Content
There is strong evidence that the efficiency of the Internet is limited by its existing TCP congestion control system. A replacement, FAST, has been shown to improve performance in high-speed networks. In order to achieve widespread acceptance and standardisation, it must also be tested in environments more typical of the existing Internet. This paper experimentally evaluates the performance of FAST over a typical access link, with bandwidths of around 0.5-3 Mbps. Links both using the DOCSIS cable modem medium access control (MAC) protocol and simple low rate links were investigated. It is shown that the random delay introduced by MAC protocol of the cable modem does not appear to interfere significantly with FAST's ability to set the congestion window size to its target. However, the cable modem does appear to introduce consistent additional delays when the link is highly, but not fully, utilised. These unexplained delays mean that a larger congestion window is required, and must be taken into account when setting FAST's parameters, notably the target queue size, alpha.