Estimating loss rates with TCP

Estimating loss rates along a network path is a problem that has received much attention within the research community. However, deriving accurate estimates of the loss rate from TCP transfers has been largely unaddressed. In this paper, we first show that using a simple count of the number of retransmissions yields inaccurate estimates of the loss rate in many cases. The mis-estimation stems from flaws in TCP's retransmission schemes that cause the protocol to spuriously retransmit data in a number of cases. Next, we develop techniques for refining the retransmission count to produce a better loss rate estimate for both Reno and SACK variants of TCP. Finally, we explore two SACK-based variants of TCP with an eye towards reducing spurious retransmits, the root cause of the mis-estimation of the loss rate. An additional benefit of reducing the number of needless retransmits is a reduction in the amount of shared network resources used to accomplish no useful work.

[1]  Hao Jiang,et al.  Passive estimation of TCP round-trip times , 2002, CCRV.

[2]  Matthew Mathis,et al.  Automatic TCP buffer tuning , 1998, SIGCOMM '98.

[3]  Sally Floyd,et al.  The NewReno Modification to TCP's Fast Recovery Algorithm , 2004, RFC.

[4]  Rajesh Krishnan,et al.  Explicit transport error notification (ETEN) for error-prone wireless and satellite networks , 2004, Comput. Networks.

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

[6]  Randy H. Katz,et al.  The Eifel algorithm: making TCP robust against spurious retransmissions , 2000, CCRV.

[7]  Janey C. Hoe.  Improving the start-up behavior of a congestion control scheme for TCP , 1996, SIGCOMM '96.

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

[9]  Mark Allman,et al.  On making TCP more robust to packet reordering , 2002, CCRV.

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

[11]  Vern Paxson,et al.  End-to-end Internet packet dynamics , 1997, SIGCOMM '97.

[12]  Sally Floyd,et al.  Identifying the tcp behavior of web servers , 2000, SIGCOMM 2000.

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

[14]  Péter Benkö,et al.  A passive method for estimating end-to-end TCP packet loss , 2002, Global Telecommunications Conference, 2002. GLOBECOM '02. IEEE.

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

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

[17]  Lili Wang,et al.  A Conservative Selective Acknowledgment (SACK)-based Loss Recovery Algorithm for TCP , 2003, RFC.

[18]  Sally Floyd,et al.  The NewReno Modification to TCP's Fast Recovery Algorithm , 2004, RFC.

[19]  Mark Allman,et al.  New techniques for making transport protocols robust to corruption-based loss , 2004, CCRV.

[20]  Rajeev Koodli,et al.  One-way Loss Pattern Sample Metrics , 2002, RFC.

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

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

[23]  Mark Allman,et al.  Measuring end-to-end bulk transfer capacity , 2001, IMW '01.

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

[25]  Sally Floyd,et al.  An Extension to the Selective Acknowledgement (SACK) Option for TCP , 2000, RFC.

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

[27]  Sally Floyd,et al.  Congestion Control Principles , 2000, RFC.

[28]  AllmanMark,et al.  New techniques for making transport protocols robust to corruption-based loss , 2004 .

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

[30]  Stefan Savage,et al.  Sting: A TCP-based Network Measurement Tool , 1999, USENIX Symposium on Internet Technologies and Systems.