A Benchmark for the Performance of Time-varying Closed-loop Flow Control with Application to TCP

TL;DR

This paper introduces a measurement-based benchmarking method for evaluating TCP's delay performance, especially for real-time applications, by analyzing end-to-end network effects and protocol stack delays.

Contribution

It presents a novel estimation technique for TCP delay performance using end-to-end measurements, accounting for protocol stack effects and enabling comparison of TCP mechanisms.

Findings

Significant delays are caused by queueing in protocol stacks even at moderate utilization.

Proper selection of TCP parameters can substantially improve delay performance.

The method provides insights for network dimensioning and server configuration to optimize real-time communication.

Abstract

Closed-loop flow control protocols, such as the prominent implementation TCP, are prevalent in the Internet, today. TCP has continuously been improved for greedy traffic sources to achieve high throughput over networks with large bandwidth delay products. Recently, the increasing use for streaming and interactive applications, such as voice and video, has shifted the focus towards its delay performance. Given the need for real-time communication of non-greedy sources via TCP, we present an estimation method for performance evaluation of closed-loop flow control protocols. We characterize an end-to-end connection by a transfer function that provides statistical service guarantees for arbitrary traffic. The estimation is based on end-to-end measurements at the application level that include all effects induced by the network and by the protocol stacks of the end systems. From our measurements, we identify different causes for delays. We show that significant delays are due to queueing in protocol stacks. Notably, this occurs even if the utilization is moderate. Using our estimation method, we compare the impact of fundamental mechanisms of TCP on delays at the application level: In detail, we analyze parameters relevant for network dimensioning, including buffer provisioning and active queue management, and parameters for server configuration, such as the congestion control algorithm. By applying our method as a benchmark, we find that a good selection can largely improve the delay performance of TCP.