Analysis and Improvement of Fairness between TCP Reno and Vegas for Deployment of TCP Vegas to the Internet

According to the past researches, a TCP Vegas version is able to achieve higher throughput than TCP Tahoe and Reno versions, which are widely used in the current Internet. However, we need to consider a migration path for TCP Vegas to be deployed in the Internet. By focusing on the situation where TCP Reno and Vegas connections share the bottleneck link, we investigate the fairness between two versions. Weconsider drop-tail and RED (Random Early Detection) as scheduling disciplines at the router buffer. From the analysis and the simulation results, we find that the fairness between TCP Reno and Vegas can not be kept at all with the drop-tail router, and the performance of TCP Vegas is much smaller than that of TCP Reno as opposed to an expectation on TCP Vegas. The RED algorithm improves the fairness to some degree, but there still be an inevitable trade-off between fairness and throughput. It is true that TCP Vegas solely can obtain high throughput,and it has a good feature of having a backward compatibility with older versions of TCP. Nevertheless, it is unlikely that a current version of TCP Vegas penetrates in the Internet as our results clearly indicate. Accordingly, we next consider two approaches to improve the fairness. The first one is to modify the congestion control algorithm of TCP Vegas, and the other is to modify the RED algorithm to detect mis-behaved connections and drop more packets from those connections. We use both of analysis and simulation experiment for evaluating the fairness, and validate the effectiveness of the proposed mechanisms.