Browse Prior Art Database

TCP Congestion Control with Appropriate Byte Counting (ABC) (RFC3465)

IP.com Disclosure Number: IPCOM000011360D
Original Publication Date: 2003-Feb-01
Included in the Prior Art Database: 2003-Feb-14
Document File: 11 page(s) / 23K

Publishing Venue

Internet Society Requests For Comment (RFCs)

Related People

M. Allman: AUTHOR

Abstract

This document proposes a small modification to the way TCP increases its congestion window. Rather than the traditional method of increasing the congestion window by a constant amount for each arriving acknowledgment, the document suggests basing the increase on the number of previously unacknowledged bytes each ACK covers. This change improves the performance of TCP, as well as closes a security hole TCP receivers can use to induce the sender into increasing the sending rate too rapidly.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 12% of the total text.

Network Working Group                                          M. Allman

Request for Comments: 3465                                  BBN/NASA GRC

Category: Experimental                                     February 2003

      TCP Congestion Control with Appropriate Byte Counting (ABC)

Status of this Memo

   This memo defines an Experimental Protocol for the Internet

   community.  It does not specify an Internet standard of any kind.

   Discussion and suggestions for improvement are requested.

   Distribution of this memo is unlimited.

Copyright Notice

   Copyright (C) The Internet Society (2003).  All Rights Reserved.

Abstract

   This document proposes a small modification to the way TCP increases

   its congestion window.  Rather than the traditional method of

   increasing the congestion window by a constant amount for each

   arriving acknowledgment, the document suggests basing the increase on

   the number of previously unacknowledged bytes each ACK covers.  This

   change improves the performance of TCP, as well as closes a security

   hole TCP receivers can use to induce the sender into increasing the

   sending rate too rapidly.

Terminology

   Much of the language in this document is taken from [RFC2581].

   The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",

   "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this

   document are to be interpreted as described in [RFC2119].

1   Introduction

   This document proposes a modification to the algorithm for increasing

   TCP's congestion window (cwnd) that improves both performance and

   security.  Rather than increasing a TCP's congestion window based on

   the number of acknowledgments (ACKs) that arrive at the data sender

   (per the current specification [RFC2581]), the congestion window is

   increased based on the number of bytes acknowledged by the arriving

   ACKs.  The algorithm improves performance by mitigating the impact of

   delayed ACKs on the growth of cwnd.  At the same time, the algorithm

   provides cwnd growth in direct relation to the probed capacity of a

Allman                        Experimental                      [Page 1]

RFC 3465            TCP Congestion Control with ABC        February 2003

   network path, therefore providing a more measured response to ACKs

   that cover only small amounts of data (less than a full segment size)

   than ACK counting.  This more appropriate cwnd growth can improve

   both performance and can prevent inappropriate cwnd growth in

   response to a misbehaving receiver.  On the other hand, in some cases

   the modified cwnd growth algorithm causes larger bursts of segments

   to be sent into the network.  In some cases this can lead to a non-

   negligible increase in the drop rate and reduced perfo...