Measuring end-to-end bulk transfer capacity

This paper provides a preliminary assessment of the effectiveness of an application layer tool that measures the Bulk Transfer Capacity (BTC) of a network path. BTC is roughly defined as the throughput that a flow using standard congestion control techniques would obtain across a given network path at a given time. We utilize the NIMI mesh of measurement hosts to compare stock BSD TCP with a new BTC measurement tool, cap. While BTC tools have been around for some time, no systematic evaluation of their accuracy with respect to standard TCP congestion control across a wide variety of network paths has been conducted. The goal of this paper is to provide such an empirical evaluation of a BTC tool and therefore assess the reliability of the measurement obtained using BTC tools.

[1]  Steven McCanne,et al.  The BSD Packet Filter: A New Architecture for User-level Packet Capture , 1993, USENIX Winter.

[2]  Srinivasan Keshav.  A control-theoretic approach to flow control , 1991, SIGCOMM 1991.

[3]  Vern Paxson,et al.  Experiences with NIMI , 2002, Proceedings 2002 Symposium on Applications and the Internet (SAINT) Workshops.

[4]  Vern Paxson,et al.  TCP Congestion Control , 1999, RFC.

[5]  Mark Allman,et al.  A Framework for Defining Empirical Bulk Transfer Capacity Metrics , 2001, RFC.

[6]  Vern Paxson,et al.  Automated packet trace analysis of TCP implementations , 1997, SIGCOMM '97.

[7]  K. Claffy,et al.  Trends in wide area IP traffic patterns - A view from Ames Internet Exchange , 2000 .

[8]  Randy H. Katz,et al.  The effects of asymmetry on TCP performance , 1997, MobiCom '97.

[9]  Vern Paxson,et al.  On estimating end-to-end network path properties , 2001, SIGCOMM LA '01.

[10]  Sally Floyd,et al.  TCP and Successive Fast Retransmits , 1995 .

[11]  Sally Floyd,et al.  Increasing TCP's Initial Window , 1998, RFC.

[12]  Robert T. Braden,et al.  Requirements for Internet Hosts - Communication Layers , 1989, RFC.

[13]  Van Jacobson,et al.  TCP Extensions for High Performance , 1992, RFC.

[14]  Craig Partridge,et al.  Packet reordering is not pathological network behavior , 1999, TNET.

[15]  Mark Allman,et al.  On the effective evaluation of TCP , 1999, CCRV.

[16]  Vern Paxson,et al.  An architecture for large-scale Internet measurement , 1998, IEEE Commun. Mag..

[17]  Sally Floyd,et al.  TCP Selective Acknowledgement Options , 1996 .

[18]  Vern Paxson,et al.  Computing TCP's Retransmission Timer , 2000, RFC.

[19]  Donald F. Towsley,et al.  Modeling TCP throughput: a simple model and its empirical validation , 1998, SIGCOMM '98.

[20]  Van Jacobson,et al.  A tool to infer characteristics of internet paths , 1997 .

[21]  Matthew Mathis,et al.  The macroscopic behavior of the TCP congestion avoidance algorithm , 1997, CCRV.

[22]  Sally Floyd,et al.  Simulation-based comparisons of Tahoe, Reno and SACK TCP , 1996, CCRV.

[23]  Mark Crovella,et al.  Measuring Bottleneck Link Speed in Packet-Switched Networks , 1996, Perform. Evaluation.

[24]  Mark Allman,et al.  An evaluation of TCP with larger initial windows , 1998, CCRV.