Extensive evaluation on the performance and behaviour of TCP congestion control protocols under varied network scenarios

Abstract

In recent decades, many TCP Congestion Control (CC) protocols have been proposed to improve the performance and reliability of TCP in various network scenarios. However, CC protocols are usually closely coupled with network conditions such as latency and packet loss. Considering that networks with different properties are common, e.g., wired/wireless LAN and Long Fat Networks (LFNs), investigating both performance and behaviors of CC protocols under varied network scenarios becomes crucial for both network management and development. In this paper, we conduct a comprehensive measurement study on the goodput, RTT, retransmission, friendliness, fairness, convergence time and stability of most widely-used CC protocols over wired LAN/WAN and wireless LAN (both 2.4GHz and 5GHz Wi-Fi). We also conduct comparative studies with respect to transmission cost, congested reverse path and bottleneck queue size in network simulator.Based on our analysis, we reveal several interesting and original observations. We found that the goodput of BBR is at least 22.5% lower than other CC protocols in wireless LAN due to insufficient pacing rate, even though it can always fully utilize the bottleneck bandwidth with low RTT in wired networks. We also observed that the total on-wire data volume of BBR is higher

An Evaluation of BBR and its variants

“…e congestion control for TCP bears such importance that it guarantees the internet can work in normal status and avoids congestion collapse happening again. Several algorithms have been proposed by making minor changes to the basic AIMD control law and to adapt it in some speci c network environments [2] [3].…”

TCP-WBQ: a backlog-queue-based congestion control mechanism for heterogeneous wireless networks

“…TCP-W + reduces cwnd and the start threshold ( ssthresh ) according to the estimated bandwidth ( EB ) rather than recklessly cutting these values 26 as loss-based variants do. Nevertheless, due to adopting additive increase in cwnd , TCP-W + , in the multi-traffic competition, occupies the bandwidth of the bottleneck link slower than the loss-based TCP variants over better-wired links 27 . This unfairness increases the latency upon the wireless link, thus enabling TCP-W + to often under-estimate the available bandwidth.…”

P4air: Increasing Fairness among Competing Congestion Control Algorithms

“…However, due to this abundance in new protocols and algorithms, it has become almost impossible to take their interactions with other protocols and algorithms into account. Consequently, even for algorithms designed with good fairness properties in mind, multiple fairness issues exist, especially when the bottleneck links are shared between flows using different congestion control algorithms or having different Round-Trip Times (RTTs) [5], [6], [8], [24], [36], [40], [53], [56], [60]. For example, classic TCP flows (using loss-based algorithms) fill the bottleneck queues (resulting in high queuing delay) and only react to the resulting packet loss.…”