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...