On Inferring TCP Behavior

Authors: Jitendra Pahdye and Sally Floyd

Complete Citation

• Jitendra Pahdye and Sally Floyd. On Inferring TCP Behavior. Proceedings of the 2001 conference on Applications, technologies, architectures, and protocols for computer communications, p.287-298, August 2001, San Diego, California, United States

Abstract

Most of the traffic in today's Internet is controlled by the Transmission Control Protocol (TCP). Hence, the performance of TCP has a significant impact on the performance of the overall Internet. TCP is a complex protocol with many user-configurable parameters and a range of different implementations. In addition, research continues to produce new developments in congestion control mechanisms and TCP options, and it is useful to trace the deployment of these new mechanisms in the Internet. As a final concern, the stability and fairness of the current Internet relies on the voluntary use of congestion control mechanisms by end hosts. Therefore it is important to test TCP implementations for conformant end-to-end congestion control. Since web traffic forms the majority of the TCP traffic, TCP implementations in today's web servers are of particular interest. We have developed a tool called TCP Behavior Inference Tool (TBIT) to characterize the TCP behavior of a remote web server. In this paper, we describe TBIT, and present results about the TCP behaviors of major web servers, obtained using this tool. We also describe the use of TBIT to detect bugs and non-compliance in TCP implementations deployed in public web servers.

Annotations

The authors developed a tool called TCP Behavior Inference Tool (TBIT) to characterize the TCP behavior of a remote web server. The TBIT process establishes and maintains a TCP connection with the remote host entirely at the user level. This user-level TCP connection can then be manipulated to extract information about the remote TCP. Here are some test results:

• Initial value of congestion window (ICW): 81% of the web servers set the ICW to two segments, while 13.8% of the servers set it to a single segment.
• Conformant congestion control (CCC): A TCP sender is expected to halve its congestion window after a packet loss. In the test, 3330 servers out of 3461 reduced their congestion window according to packet losses.
• Response to selective acknowledgments: The majority of servers (around 60%) are not SACK-capable.

Related Work

Alberto Medina , Mark Allman , Sally Floyd, Measuring the Evolution of Transport Protocols in the Internet, ACM SIGCOMM Computer Communication Review, v.35 n.2, April 2005

• The distribution of minimum MSS sizes: nearly all servers will accept an MSS as small as 128 bytes, with many servers supporting MSS sizes of 32 and 64 bytes. However, the lower the MSS size, the higher percentage of reordered segments.

-- YingxinJiang - 09 Jul 2007