Enhancing TCP's Loss Recovery Using Limited Transmit
RFC 3042
Network Working Group M. Allman
Request for Comments: 3042 NASA GRC/BBN
Category: Standards Track H. Balakrishnan
MIT
S. Floyd
ACIRI
January 2001
Enhancing TCP's Loss Recovery Using Limited Transmit
Status of this Memo
This document specifies an Internet standards track protocol for the
Internet community, and requests discussion and suggestions for
improvements. Please refer to the current edition of the "Internet
Official Protocol Standards" (STD 1) for the standardization state
and status of this protocol. Distribution of this memo is unlimited.
Copyright Notice
Copyright (C) The Internet Society (2001). All Rights Reserved.
Abstract
This document proposes a new Transmission Control Protocol (TCP)
mechanism that can be used to more effectively recover lost segments
when a connection's congestion window is small, or when a large
number of segments are lost in a single transmission window. The
"Limited Transmit" algorithm calls for sending a new data segment in
response to each of the first two duplicate acknowledgments that
arrive at the sender. Transmitting these segments increases the
probability that TCP can recover from a single lost segment using the
fast retransmit algorithm, rather than using a costly retransmission
timeout. Limited Transmit can be used both in conjunction with, and
in the absence of, the TCP selective acknowledgment (SACK) mechanism.
1 Introduction
A number of researchers have observed that TCP's loss recovery
strategies do not work well when the congestion window at a TCP
sender is small. This can happen, for instance, because there is
only a limited amount of data to send, or because of the limit
imposed by the receiver-advertised window, or because of the
constraints imposed by end-to-end congestion control over a
connection with a small bandwidth-delay product
[Riz96,Mor97,BPS+98,Bal98,LK98]. When a TCP detects a missing
segment, it enters a loss recovery phase using one of two methods.
Allman, et al. Standards Track [Page 1]
RFC 3042 Enhancing TCP Loss Recovery January 2001
First, if an acknowledgment (ACK) for a given segment is not received
in a certain amount of time a retransmission timeout occurs and the
segment is resent [RFC793,PA00]. Second, the "Fast Retransmit"
algorithm resends a segment when three duplicate ACKs arrive at the
sender [Jac88,RFC2581]. However, because duplicate ACKs from the
receiver are also triggered by packet reordering in the Internet, the
TCP sender waits for three duplicate ACKs in an attempt to
disambiguate segment loss from packet reordering. Once in a loss
recovery phase, a number of techniques can be used to retransmit lost
segments, including slow start-based recovery or Fast Recovery
[RFC2581], NewReno [RFC2582], and loss recovery based on selective
acknowledgments (SACKs) [RFC2018,FF96].
TCP's retransmission timeout (RTO) is based on measured round-trip
times (RTT) between the sender and receiver, as specified in [PA00].
To prevent spurious retransmissions of segments that are only delayed
and not lost, the minimum RTO is conservatively chosen to be 1
second. Therefore, it behooves TCP senders to detect and recover
from as many losses as possible without incurring a lengthy timeout
when the connection remains idle. However, if not enough duplicate
ACKs arrive from the receiver, the Fast Retransmit algorithm is never
triggered---this situation occurs when the congestion window is small
or if a large number of segments in a window are lost. For instance,
consider a congestion window (cwnd) of three segments. If one
segment is dropped by the network, then at most two duplicate ACKs
will arrive at the sender. Since three duplicate ACKs are required
to trigger Fast Retransmit, a timeout will be required to resend the
dropped packet.
[BPS+97] found that roughly 56% of retransmissions sent by a busy web
server were sent after the RTO expires, while only 44% were handled
by Fast Retransmit. In addition, only 4% of the RTO-based
retransmissions could have been avoided with SACK, which of course
has to continue to disambiguate reordering from genuine loss. In
contrast, using the technique outlined in this document and in
[Bal98], 25% of the RTO-based retransmissions in that dataset would
have likely been avoided.
The next section of this document outlines small changes to TCP
senders that will decrease the reliance on the retransmission timer,
Show full document text