draft-ietf-tsvwg-newreno-02.txt   rfc3782.txt 
Internet Engineering Task Force S. Floyd Network Working Group S. Floyd
INTERNET DRAFT ICSI Request for Comments: 3782 ICSI
draft-ietf-tsvwg-newreno-02.txt T. Henderson Obsoletes: 2582 T. Henderson
Boeing Category: Standards Track Boeing
A. Gurtov A. Gurtov
U. Helsinki TeliaSonera
November 2003 April 2004
The NewReno Modification to TCP's Fast Recovery Algorithm The NewReno Modification to TCP's Fast Recovery Algorithm
Status of this Memo Status of this Memo
This document is an Internet-Draft and is in full conformance with This document specifies an Internet standards track protocol for the
all provisions of Section 10 of RFC2026. Internet community, and requests discussion and suggestions for
improvements. Please refer to the current edition of the "Internet
Internet-Drafts are working documents of the Internet Engineering Official Protocol Standards" (STD 1) for the standardization state
Task Force (IETF), its areas, and its working groups. Note that and status of this protocol. Distribution of this memo is unlimited.
other groups may also distribute working documents as Internet-
Drafts.
Internet-Drafts are draft documents valid for a maximum of six months
and may be updated, replaced, or obsoleted by other documents at any
time. It is inappropriate to use Internet-Drafts as reference
material or to cite them other than as "work in progress."
The list of current Internet-Drafts can be accessed at Copyright Notice
http://www.ietf.org/ietf/1id-abstracts.txt
The list of Internet-Draft Shadow Directories can be accessed at Copyright (C) The Internet Society (2004). All Rights Reserved.
http://www.ietf.org/shadow.html.
Abstract Abstract
RFC 2581 [RFC2581] documents the following four intertwined TCP The purpose of this document is to advance NewReno TCP's Fast
congestion control algorithms: Slow Start, Congestion Avoidance, Fast Retransmit and Fast Recovery algorithms in RFC 2582 from Experimental
Retransmit, and Fast Recovery. RFC 2581 [RFC2581] explicitly allows to Standards Track status.
certain modifications of these algorithms, including modifications
that use the TCP Selective Acknowledgement (SACK) option
[RFC2018,RFC3517], and modifications that respond to "partial
acknowledgments" (ACKs which cover new data, but not all the data
outstanding when loss was detected) in the absence of SACK. The
NewReno mechanism uses an algorithm for responding to partial
acknowledgments that was first proposed by Janey Hoe in [Hoe95].
RFC 2582 [RFC2582] specified the NewReno mechanisms as Experimental
in 1999. This document is a small revision of RFC 2582 intended to
advance the NewReno mechanisms to Proposed Standard. RFC 2581 notes
that the Fast Retransmit/Fast Recovery algorithm specified in that
document does not recover very efficiently from multiple losses in a
single flight of packets, and that RFC 2582 contains one set of
modifications to address this problem.
NOTE TO THE RFC EDITOR: PLEASE REMOVE THIS SECTION UPON PUBLICATION.
Changes from draft-ietf-tsvwg-newreno-01.txt:
* Some improvements in phrasing suggested by Mark Allman.
Changes from draft-ietf-tsvwg-newreno-00.txt:
* In Section 8, added a cautionary note about using the duplicate
acknowledgment counter as a flag for whether Fast Recovery is in
effect.
* In Section 8, added a note about pulling along "recover" with
"snd_una" when Fast Recovery is not in effect.
* Added a discussion in Section 6 about heuristics for distinguishing
between a retransmitted packet that was dropped, and three duplicate
acknowledgements simply from the unnecessary retransmission of three
packets.
* Added more text and examples for comparing the Impatient and the
Slow-but-Steady variants.
* In Section 8, added a cautionary note saying that when the sender
is not in Fast Retransmit, the sender should not use the Fast
Recovery response to multiple duplicate acknowledgements.
Changes from draft-floyd-newreno-00.txt:
* In Section 8 on "Implementation issues for the data sender",
mentioned alternate methods for limiting bursts when exiting Fast
Recovery.
* Changed draft from draft-floyd-newreno to draft-ietf-tsvwg-newreno
Changes from RFC 2582:
* Rephrasing and rearrangements of the text.
* RFC 2582 described the Careful and Less Careful variants of
NewReno, along with a default version that was neither Careful nor
Less Careful, and recommended the Careful variant. This document
only specifies the Careful version.
* RFC 2582 used two separate variables, "send_high" and "recover",
and this document has merged them into a single variable "recover".
* Added sections on "Comparisons between Reno and NewReno TCP", and
on "Changes relative to RFC 2582". The section on "Comparisons
between Reno and NewReno TCP" includes a discussion of the one area
where NewReno is known to perform worse than Reno or SACK, and that
is in the response to reordering.
* Moved all of the discussions of the Impatient and Slow-but-Steady
variants to one place, and recommended the Impatient variant (as in
the default version in RFC 2582).
* Added a section on Implementation issues for the data sender,
mentioning maxburst_.
* Added a paragraph about differences between RFC 2582 and [FF96].
END OF NOTE TO RFC EDITOR The main change in this document relative to RFC 2582 is to specify
the Careful variant of NewReno's Fast Retransmit and Fast Recovery
algorithms. The base algorithm described in RFC 2582 did not attempt
to avoid unnecessary multiple Fast Retransmits that can occur after a
timeout. However, RFC 2582 also defined "Careful" and "Less Careful"
variants that avoid these unnecessary Fast Retransmits, and
recommended the Careful variant. This document specifies the
previously-named "Careful" variant as the basic version of NewReno
TCP.
1. Introduction 1. Introduction
For the typical implementation of the TCP Fast Recovery algorithm For the typical implementation of the TCP Fast Recovery algorithm
described in [RFC2581] (first implemented in the 1990 BSD Reno described in [RFC2581] (first implemented in the 1990 BSD Reno
release, and referred to as the Reno algorithm in [FF96]), the TCP release, and referred to as the Reno algorithm in [FF96]), the TCP
data sender only retransmits a packet after a retransmit timeout has data sender only retransmits a packet after a retransmit timeout has
occurred, or after three duplicate acknowledgements have arrived occurred, or after three duplicate acknowledgements have arrived
triggering the Fast Retransmit algorithm. A single retransmit triggering the Fast Retransmit algorithm. A single retransmit
timeout might result in the retransmission of several data packets, timeout might result in the retransmission of several data packets,
but each invocation of the Fast Retransmit algorithm in RFC 2581 but each invocation of the Fast Retransmit algorithm in RFC 2581
leads to the retransmission of only a single data packet. leads to the retransmission of only a single data packet.
Problems can arise, therefore, when multiple packets have been Problems can arise, therefore, when multiple packets are dropped from
dropped from a single window of data and the Fast Retransmit and Fast a single window of data and the Fast Retransmit and Fast Recovery
Recovery algorithms are invoked. In this case, if the SACK option is algorithms are invoked. In this case, if the SACK option is
available, the TCP sender has the information to make intelligent available, the TCP sender has the information to make intelligent
decisions about which packets to retransmit and which packets not to decisions about which packets to retransmit and which packets not to
retransmit during Fast Recovery. This document applies only for TCP retransmit during Fast Recovery. This document applies only for TCP
connections that are unable to use the TCP Selective Acknowledgement connections that are unable to use the TCP Selective Acknowledgement
(SACK) option, either because the option is not locally supported or (SACK) option, either because the option is not locally supported or
because the TCP peer did not indicate a willingness to use SACK. because the TCP peer did not indicate a willingness to use SACK.
In the absence of SACK, there is little information available to the In the absence of SACK, there is little information available to the
TCP sender in making retransmission decisions during Fast Recovery. TCP sender in making retransmission decisions during Fast Recovery.
From the three duplicate acknowledgements, the sender infers a packet From the three duplicate acknowledgements, the sender infers a packet
loss, and retransmits the indicated packet. After this, the data loss, and retransmits the indicated packet. After this, the data
sender could receive additional duplicate acknowledgements, as the sender could receive additional duplicate acknowledgements, as the
data receiver acknowledges additional data packets that were already data receiver acknowledges additional data packets that were already
in flight when the sender entered Fast Retransmit. in flight when the sender entered Fast Retransmit.
In the case of multiple packets dropped from a single window of data, In the case of multiple packets dropped from a single window of data,
the first new information available to the sender comes when the the first new information available to the sender comes when the
sender receives an acknowledgement for the retransmitted packet (that sender receives an acknowledgement for the retransmitted packet (that
is, the packet retransmitted when Fast Retransmit was first entered). is, the packet retransmitted when Fast Retransmit was first entered).
skipping to change at page 4, line 15 skipping to change at page 2, line 39
From the three duplicate acknowledgements, the sender infers a packet From the three duplicate acknowledgements, the sender infers a packet
loss, and retransmits the indicated packet. After this, the data loss, and retransmits the indicated packet. After this, the data
sender could receive additional duplicate acknowledgements, as the sender could receive additional duplicate acknowledgements, as the
data receiver acknowledges additional data packets that were already data receiver acknowledges additional data packets that were already
in flight when the sender entered Fast Retransmit. in flight when the sender entered Fast Retransmit.
In the case of multiple packets dropped from a single window of data, In the case of multiple packets dropped from a single window of data,
the first new information available to the sender comes when the the first new information available to the sender comes when the
sender receives an acknowledgement for the retransmitted packet (that sender receives an acknowledgement for the retransmitted packet (that
is, the packet retransmitted when Fast Retransmit was first entered). is, the packet retransmitted when Fast Retransmit was first entered).
If there had been a single packet drop and no reordering, then the If there is a single packet drop and no reordering, then the
acknowledgement for this packet will acknowledge all of the packets acknowledgement for this packet will acknowledge all of the packets
transmitted before Fast Retransmit was entered. However, when there transmitted before Fast Retransmit was entered. However, if there
were multiple packet drops, then the acknowledgement for the are multiple packet drops, then the acknowledgement for the
retransmitted packet will acknowledge some but not all of the packets retransmitted packet will acknowledge some but not all of the packets
transmitted before the Fast Retransmit. We call this acknowledgement transmitted before the Fast Retransmit. We call this acknowledgement
a partial acknowledgment. a partial acknowledgment.
Along with several other suggestions, [Hoe95] suggested that during Along with several other suggestions, [Hoe95] suggested that during
Fast Recovery the TCP data sender responds to a partial Fast Recovery the TCP data sender responds to a partial
acknowledgment by inferring that the next in-sequence packet has been acknowledgment by inferring that the next in-sequence packet has been
lost, and retransmitting that packet. This document describes a lost, and retransmitting that packet. This document describes a
modification to the Fast Recovery algorithm in RFC 2581 that modification to the Fast Recovery algorithm in RFC 2581 that
incorporates a response to partial acknowledgements received during incorporates a response to partial acknowledgements received during
Fast Recovery. We call this modified Fast Recovery algorithm Fast Recovery. We call this modified Fast Recovery algorithm
NewReno, because it is a slight but significant variation of the NewReno, because it is a slight but significant variation of the
basic Reno algorithm in RFC 2581. This document does not discuss the basic Reno algorithm in RFC 2581. This document does not discuss the
other suggestions in [Hoe95] and [Hoe96], such as a change to the other suggestions in [Hoe95] and [Hoe96], such as a change to the
ssthresh parameter during Slow-Start, or the proposal to send a new ssthresh parameter during Slow-Start, or the proposal to send a new
packet for every two duplicate acknowledgements during Fast Recovery. packet for every two duplicate acknowledgements during Fast Recovery.
The version of NewReno in this document also draws on other The version of NewReno in this document also draws on other
discussions of NewReno in the literature [LM97]. discussions of NewReno in the literature [LM97, Hen98].
We do not claim that the NewReno version of Fast Recovery described We do not claim that the NewReno version of Fast Recovery described
here is an optimal modification of Fast Recovery for responding to here is an optimal modification of Fast Recovery for responding to
partial acknowledgements, for TCP connections that are unable to use partial acknowledgements, for TCP connections that are unable to use
SACK. Based on our experiences with the NewReno modification in the SACK. Based on our experiences with the NewReno modification in the
NS simulator [NS] and with numerous implementations of NewReno, we NS simulator [NS] and with numerous implementations of NewReno, we
believe that this modification improves the performance of the Fast believe that this modification improves the performance of the Fast
Retransmit and Fast Recovery algorithms in a wide variety of Retransmit and Fast Recovery algorithms in a wide variety of
scenarios. scenarios.
2. Terminology and Definitions 2. Terminology and Definitions
In this document, the key words "MUST", "MUST NOT", "REQUIRED", In this document, the key words "MUST", "MUST NOT", "REQUIRED",
"SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY",
and "OPTIONAL" are to be interpreted as described in BCP 14, RFC 2119 and "OPTIONAL" are to be interpreted as described in BCP 14, RFC 2119
and indicate requirement levels for compliant TCP implementations [RFC2119]. This RFC indicates requirement levels for compliant TCP
implementing the NewReno Fast Retransmit and Fast Recovery algorithms implementations implementing the NewReno Fast Retransmit and Fast
described in this document. Recovery algorithms described in this document.
This document assumes that the reader is familiar with the terms This document assumes that the reader is familiar with the terms
SENDER MAXIMUM SEGMENT SIZE (SMSS), CONGESTION WINDOW (cwnd), and SENDER MAXIMUM SEGMENT SIZE (SMSS), CONGESTION WINDOW (cwnd), and
FLIGHT SIZE (FlightSize) defined in [RFC2581]. FLIGHT SIZE is FLIGHT SIZE (FlightSize) defined in [RFC2581]. FLIGHT SIZE is
defined as in [RFC2581] as follows: defined as in [RFC2581] as follows:
FLIGHT SIZE: FLIGHT SIZE:
The amount of data that has been sent but not yet acknowledged. The amount of data that has been sent but not yet acknowledged.
3. The Fast Retransmit and Fast Recovery Algorithms in NewReno 3. The Fast Retransmit and Fast Recovery Algorithms in NewReno
skipping to change at page 5, line 44 skipping to change at page 4, line 20
acknowledgement in step 5, and in modifications to step 1 and the acknowledgement in step 5, and in modifications to step 1 and the
addition of step 6 for avoiding multiple Fast Retransmits caused by addition of step 6 for avoiding multiple Fast Retransmits caused by
the retransmission of packets already received by the receiver. the retransmission of packets already received by the receiver.
The algorithm specified in this document uses a variable "recover", The algorithm specified in this document uses a variable "recover",
whose initial value is the initial send sequence number. whose initial value is the initial send sequence number.
1) Three duplicate ACKs: 1) Three duplicate ACKs:
When the third duplicate ACK is received and the sender is not When the third duplicate ACK is received and the sender is not
already in the Fast Recovery procedure, check to see if the already in the Fast Recovery procedure, check to see if the
Cumulative Acknowledgement field covers more than "recover". Cumulative Acknowledgement field covers more than "recover". If
If so, go to Step 1A. Otherwise, go to Step 1B. so, go to Step 1A. Otherwise, go to Step 1B.
1A) Invoking Fast Retransmit: 1A) Invoking Fast Retransmit:
If so, then set ssthresh to no more than the value given in If so, then set ssthresh to no more than the value given in
equation 1 below. (This is equation 3 from [RFC2581]). equation 1 below. (This is equation 3 from [RFC2581]).
ssthresh = max (FlightSize / 2, 2*SMSS) (1) ssthresh = max (FlightSize / 2, 2*SMSS) (1)
In addition, record the highest sequence number transmitted in In addition, record the highest sequence number transmitted in
the variable "recover", and go to Step 2. the variable "recover", and go to Step 2.
1B) Not invoking Fast Retransmit: 1B) Not invoking Fast Retransmit:
Do not enter the Fast Retransmit and Fast Recovery procedure. Do not enter the Fast Retransmit and Fast Recovery procedure. In
In particular, do not change ssthresh, do not go to Step 2 to particular, do not change ssthresh, do not go to Step 2 to
retransmit the "lost" segment, and do not execute Step 3 upon retransmit the "lost" segment, and do not execute Step 3 upon
subsequent duplicate ACKs. subsequent duplicate ACKs.
2) Entering Fast Retransmit: 2) Entering Fast Retransmit:
Retransmit the lost segment and set cwnd to ssthresh plus 3*SMSS. Retransmit the lost segment and set cwnd to ssthresh plus 3*SMSS.
This artificially "inflates" the congestion window by the number This artificially "inflates" the congestion window by the number
of segments (three) that have left the network and which the of segments (three) that have left the network and the receiver
receiver has buffered. has buffered.
3) Fast Recovery: 3) Fast Recovery:
For each additional duplicate ACK received while in Fast For each additional duplicate ACK received while in Fast
Recovery, increment cwnd by SMSS. This artificially inflates Recovery, increment cwnd by SMSS. This artificially inflates the
the congestion window in order to reflect the additional segment congestion window in order to reflect the additional segment that
that has left the network. has left the network.
4) Fast Recovery, continued: 4) Fast Recovery, continued:
Transmit a segment, if allowed by the new value of cwnd and the Transmit a segment, if allowed by the new value of cwnd and the
receiver's advertised window. receiver's advertised window.
5) When an ACK arrives that acknowledges new data, this ACK could be 5) When an ACK arrives that acknowledges new data, this ACK could be
the acknowledgment elicited by the retransmission from step 2, or the acknowledgment elicited by the retransmission from step 2, or
elicited by a later retransmission. elicited by a later retransmission.
Full acknowledgements: Full acknowledgements:
If this ACK acknowledges all of the data up to and including If this ACK acknowledges all of the data up to and including
"recover", then the ACK acknowledges all the intermediate "recover", then the ACK acknowledges all the intermediate
segments sent between the original transmission of the lost segments sent between the original transmission of the lost
segment and the receipt of the third duplicate ACK. Set cwnd to segment and the receipt of the third duplicate ACK. Set cwnd to
either (1) min (ssthresh, FlightSize + SMSS); or (2) ssthresh, either (1) min (ssthresh, FlightSize + SMSS) or (2) ssthresh,
where ssthresh is the value set in step 1; this is termed where ssthresh is the value set in step 1; this is termed
"deflating" the window. (We note that "FlightSize" in step 1 "deflating" the window. (We note that "FlightSize" in step 1
referred to the amount of data outstanding in step 1, when Fast referred to the amount of data outstanding in step 1, when Fast
Recovery was entered, while "FlightSize" in step 5 refers to the Recovery was entered, while "FlightSize" in step 5 refers to the
amount of data outstanding in step 5, when Fast Recovery is amount of data outstanding in step 5, when Fast Recovery is
exited.) If the second option is selected, the implementation exited.) If the second option is selected, the implementation is
is encouraged to take measures to avoid a possible burst of encouraged to take measures to avoid a possible burst of data, in
data, in case the amount of data outstanding in the network was case the amount of data outstanding in the network is much less
much less than the new congestion window allows. A simple than the new congestion window allows. A simple mechanism is to
mechanism is to limit the number of data packets that can be limit the number of data packets that can be sent in response to
sent in response to a single acknowledgement. (This is known a single acknowledgement; this is known as "maxburst_" in the NS
as "maxburst_" in the NS simulator). Exit the Fast Recovery simulator. Exit the Fast Recovery procedure.
procedure.
Partial acknowledgements: Partial acknowledgements:
If this ACK does *not* acknowledge all of the data up to and If this ACK does *not* acknowledge all of the data up to and
including "recover", then this is a partial ACK. In this case, including "recover", then this is a partial ACK. In this case,
retransmit the first unacknowledged segment. Deflate the retransmit the first unacknowledged segment. Deflate the
congestion window by the amount of new data acknowledged by the congestion window by the amount of new data acknowledged by the
cumulative acknowledgement field. If the partial ACK cumulative acknowledgement field. If the partial ACK
acknowledges at least one SMSS of new data, then add back SMSS acknowledges at least one SMSS of new data, then add back SMSS
bytes to the congestion window. As in Step 3, this artificially bytes to the congestion window. As in Step 3, this artificially
inflates the congestion window in order to reflect the additional inflates the congestion window in order to reflect the additional
segment that has left the network. Send a new segment if segment that has left the network. Send a new segment if
permitted by the new value of cwnd. This "partial window permitted by the new value of cwnd. This "partial window
deflation" attempts to ensure that, when Fast Recovery eventually deflation" attempts to ensure that, when Fast Recovery eventually
ends, approximately ssthresh amount of data will be outstanding ends, approximately ssthresh amount of data will be outstanding
in the network. Do not exit the Fast Recovery procedure (i.e., in the network. Do not exit the Fast Recovery procedure (i.e.,
if any duplicate ACKs subsequently arrive, execute Steps 3 and if any duplicate ACKs subsequently arrive, execute Steps 3 and 4
4 above). above).
For the first partial ACK that arrives during Fast Recovery, also For the first partial ACK that arrives during Fast Recovery, also
reset the retransmit timer. Timer management is discussed in reset the retransmit timer. Timer management is discussed in
more detail in Section 4. more detail in Section 4.
6) Retransmit timeouts: 6) Retransmit timeouts:
After a retransmit timeout, record the highest sequence number After a retransmit timeout, record the highest sequence number
transmitted in the variable "recover" and exit the Fast transmitted in the variable "recover" and exit the Fast Recovery
Recovery procedure if applicable. procedure if applicable.
Step 1 specifies a check that the Cumulative Acknowledgement field Step 1 specifies a check that the Cumulative Acknowledgement field
covers more than "recover". Because the acknowledgement field covers more than "recover". Because the acknowledgement field
contains the sequence number that the sender next expects to receive, contains the sequence number that the sender next expects to receive,
the acknowledgement "ack_number" covers more than "recover" when: the acknowledgement "ack_number" covers more than "recover" when:
ack_number - 1 > recover. ack_number - 1 > recover;
i.e., at least one byte more of data is acknowledged beyond the
highest byte that was outstanding when Fast Retransmit was last
entered.
Note that in Step 5, the congestion window is deflated after a Note that in Step 5, the congestion window is deflated after a
partial acknowledgement is received. The congestion window was partial acknowledgement is received. The congestion window was
likely to have been inflated considerably when the partial likely to have been inflated considerably when the partial
acknowledgement was received. In addition, depending on the original acknowledgement was received. In addition, depending on the original
pattern of packet losses, the partial acknowledgement might pattern of packet losses, the partial acknowledgement might
acknowledge nearly a window of data. In this case, if the congestion acknowledge nearly a window of data. In this case, if the congestion
window was not deflated, the data sender might be able to send nearly window was not deflated, the data sender might be able to send nearly
a window of data back-to-back. a window of data back-to-back.
skipping to change at page 8, line 14 skipping to change at page 6, line 46
current standard is RFC 2581, which specifies a threshold of three current standard is RFC 2581, which specifies a threshold of three
duplicate acknowledgements. duplicate acknowledgements.
As a final note, we would observe that in the absence of the SACK As a final note, we would observe that in the absence of the SACK
option, the data sender is working from limited information. When option, the data sender is working from limited information. When
the issue of recovery from multiple dropped packets from a single the issue of recovery from multiple dropped packets from a single
window of data is of particular importance, the best alternative window of data is of particular importance, the best alternative
would be to use the SACK option. would be to use the SACK option.
4. Resetting the Retransmit Timer in Response to Partial 4. Resetting the Retransmit Timer in Response to Partial
Acknowledgements. Acknowledgements
One possible variant to the response to partial acknowledgements One possible variant to the response to partial acknowledgements
specified in Section 3 concerns when to reset the retransmit timer specified in Section 3 concerns when to reset the retransmit timer
after a partial acknowledgement. The algorithm in Section 3, Step 5, after a partial acknowledgement. The algorithm in Section 3, Step 5,
resets the retransmit timer only after the first partial ACK. In resets the retransmit timer only after the first partial ACK. In
this case, if a large number of packets were dropped from a window of this case, if a large number of packets were dropped from a window of
data, the TCP data sender's retransmit timer will ultimately expire, data, the TCP data sender's retransmit timer will ultimately expire,
and the TCP data sender will invoke Slow-Start. (This is illustrated and the TCP data sender will invoke Slow-Start. (This is illustrated
on page 12 of [F98].) We call this the Impatient variant of NewReno. on page 12 of [F98].) We call this the Impatient variant of NewReno.
We note that the Impatient variant in Section 3 doesn't follow the We note that the Impatient variant in Section 3 doesn't follow the
skipping to change at page 8, line 36 skipping to change at page 7, line 19
after every packet transmission or retransmission [RFC2988, Step after every packet transmission or retransmission [RFC2988, Step
5.1]. 5.1].
In contrast, the NewReno simulations in [FF96] illustrate the In contrast, the NewReno simulations in [FF96] illustrate the
algorithm described above with the modification that the retransmit algorithm described above with the modification that the retransmit
timer is reset after each partial acknowledgement. We call this the timer is reset after each partial acknowledgement. We call this the
Slow-but-Steady variant of NewReno. In this case, for a window with Slow-but-Steady variant of NewReno. In this case, for a window with
a large number of packet drops, the TCP data sender retransmits at a large number of packet drops, the TCP data sender retransmits at
most one packet per roundtrip time. (This behavior is illustrated in most one packet per roundtrip time. (This behavior is illustrated in
the New-Reno TCP simulation of Figure 5 in [FF96], and on page 11 of the New-Reno TCP simulation of Figure 5 in [FF96], and on page 11 of
[F98]. [F98]).
When N packets have been dropped from a window of data for a large When N packets have been dropped from a window of data for a large
value of N, the Slow-but-Steady variant can remain in Fast Recovery value of N, the Slow-but-Steady variant can remain in Fast Recovery
for N round-trip times, retransmitting one more dropped packet each for N round-trip times, retransmitting one more dropped packet each
round-trip time; for these scenarios, the Impatient variant gives a round-trip time; for these scenarios, the Impatient variant gives a
faster recovery and better performance. The tests "ns test-suite- faster recovery and better performance. The tests "ns test-suite-
newreno.tcl impatient1" and "ns test-suite-newreno.tcl slow1" in the newreno.tcl impatient1" and "ns test-suite-newreno.tcl slow1" in the
NS simulator illustrate such a scenario, where the Impatient variant NS simulator illustrate such a scenario, where the Impatient variant
performs better than the Slow-but-Steady variant. The Impatient performs better than the Slow-but-Steady variant. The Impatient
variant can be particularly important for TCP connections with large variant can be particularly important for TCP connections with large
congestion windows, as illustrated by the tests "ns test-suite- congestion windows, as illustrated by the tests "ns test-suite-
newreno.tcl impatient4" and "ns test-suite-newreno.tcl slow4" in the newreno.tcl impatient4" and "ns test-suite-newreno.tcl slow4" in the
NS simulator. NS simulator.
One can also construct scenarios where the Slow-but-Steady variant One can also construct scenarios where the Slow-but-Steady variant
gives better performance than the Impatient variant. As an example, gives better performance than the Impatient variant. As an example,
this occurs then only a small number of packets are dropped, the RTO this occurs when only a small number of packets are dropped, the RTO
is sufficiently small that the retransmit timer expires, and is sufficiently small that the retransmit timer expires, and
performance would have been better without a retransmit timeout. The performance would have been better without a retransmit timeout. The
tests "ns test-suite-newreno.tcl impatient2" and "ns test-suite- tests "ns test-suite-newreno.tcl impatient2" and "ns test-suite-
newreno.tcl slow2" in the NS simulator illustrate such a scenario. newreno.tcl slow2" in the NS simulator illustrate such a scenario.
The Slow-but-Steady variant can also achieve higher goodput than the The Slow-but-Steady variant can also achieve higher goodput than the
Impatient variant, by avoiding unnecessary retransmissions. This Impatient variant, by avoiding unnecessary retransmissions. This
could be of special interest for cellular links, where every could be of special interest for cellular links, where every
transmission costs battery power and money. The tests "ns test- transmission costs battery power and money. The tests "ns test-
suite-newreno.tcl impatient3" and "ns test-suite-newreno.tcl slow3" suite-newreno.tcl impatient3" and "ns test-suite-newreno.tcl slow3"
skipping to change at page 10, line 7 skipping to change at page 8, line 39
times to recover from N losses [Hoe96]. However, in the absence of times to recover from N losses [Hoe96]. However, in the absence of
SACK, recovering as quickly as slow-start introduces the likelihood SACK, recovering as quickly as slow-start introduces the likelihood
of unnecessarily retransmitting packets, and this could significantly of unnecessarily retransmitting packets, and this could significantly
complicate the recovery mechanisms. complicate the recovery mechanisms.
We note that the response to partial acknowledgements specified in We note that the response to partial acknowledgements specified in
Section 3 of this document and in RFC 2582 differs from the response Section 3 of this document and in RFC 2582 differs from the response
in [FF96], even though both approaches only retransmit one packet in in [FF96], even though both approaches only retransmit one packet in
response to a partial acknowledgement. Step 5 of Section 3 specifies response to a partial acknowledgement. Step 5 of Section 3 specifies
that the TCP sender responds to a partial ACK by deflating the that the TCP sender responds to a partial ACK by deflating the
congestion window by the amount of new data acknowledged, then adding congestion window by the amount of new data acknowledged, adding back
back SMSS bytes if the partial ACK acknowledges at least SMSS bytes SMSS bytes if the partial ACK acknowledges at least SMSS bytes of new
of new data, and sending a new segment if permitted by the new value data, and sending a new segment if permitted by the new value of
of cwnd. Thus, only one previously-sent packet is retransmitted in cwnd. Thus, only one previously-sent packet is retransmitted in
response to each partial acknowledgement, but additional new packets response to each partial acknowledgement, but additional new packets
might be transmitted as well, depending on the amount of new data might be transmitted as well, depending on the amount of new data
acknowledged by the partial acknowledgement. In contrast, the acknowledged by the partial acknowledgement. In contrast, the
variant of NewReno illustrated in [FF96] simply set the congestion variant of NewReno illustrated in [FF96] simply set the congestion
window to ssthresh when a partial acknowledgement was received. The window to ssthresh when a partial acknowledgement was received. The
approach in [FF96] is more conservative, and does not attempt to approach in [FF96] is more conservative, and does not attempt to
accurately track the actual number of outstanding packets after a accurately track the actual number of outstanding packets after a
partial acknowledgement is received. While either of these partial acknowledgement is received. While either of these
approaches gives acceptable performance, the variant specified in approaches gives acceptable performance, the variant specified in
Section 3 recovers more smoothly when multiple packets are dropped Section 3 recovers more smoothly when multiple packets are dropped
from a window of data. (The [FF96] behavior can be seen in the NS from a window of data. (The [FF96] behavior can be seen in the NS
simulator by setting the variable "partial_window_deflation_" for simulator by setting the variable "partial_window_deflation_" for
"Agent/TCP/Newreno" to 0, and the behavior specified in Section 3 is "Agent/TCP/Newreno" to 0; the behavior specified in Section 3 is
achieved by setting "partial_window_deflation_" to 1.) achieved by setting "partial_window_deflation_" to 1.)
6. Avoiding Multiple Fast Retransmits 6. Avoiding Multiple Fast Retransmits
This section describes the motivation for the sender's state variable This section describes the motivation for the sender's state variable
"recover", and discusses possible heuristics for distinguishing "recover", and discusses possible heuristics for distinguishing
between a retransmitted packet that was dropped, and three duplicate between a retransmitted packet that was dropped, and three duplicate
acknowledgements simply from the unnecessary retransmission of three acknowledgements from the unnecessary retransmission of three
packets. packets.
In the absence of the SACK option or timestamps, a duplicate In the absence of the SACK option or timestamps, a duplicate
acknowledgement carries no information to identify the data packet or acknowledgement carries no information to identify the data packet or
packets at the TCP data receiver that triggered that duplicate packets at the TCP data receiver that triggered that duplicate
acknowledgement. In this case, the TCP data sender is unable to acknowledgement. In this case, the TCP data sender is unable to
distinguish between a duplicate acknowledgement that results from a distinguish between a duplicate acknowledgement that results from a
lost or delayed data packet, and a duplicate acknowledgement that lost or delayed data packet, and a duplicate acknowledgement that
results from the sender's unnecessary retransmission of a data packet results from the sender's unnecessary retransmission of a data packet
that had already been received at the TCP data receiver. Because of that had already been received at the TCP data receiver. Because of
skipping to change at page 11, line 10 skipping to change at page 9, line 41
the performance problems caused by multiple Fast Retransmits are the performance problems caused by multiple Fast Retransmits are
relatively minor compared to the potential problems with Tahoe TCP, relatively minor compared to the potential problems with Tahoe TCP,
which does not implement Fast Recovery. Nevertheless, unnecessary which does not implement Fast Recovery. Nevertheless, unnecessary
Fast Retransmits can occur with Reno TCP unless some explicit Fast Retransmits can occur with Reno TCP unless some explicit
mechanism is added to avoid this, such as the use of the "recover" mechanism is added to avoid this, such as the use of the "recover"
variable. (This modification is called "bugfix" in [F98], and is variable. (This modification is called "bugfix" in [F98], and is
illustrated on pages 7 and 9 of that document. Unnecessary Fast illustrated on pages 7 and 9 of that document. Unnecessary Fast
Retransmits for Reno without "bugfix" is illustrated on page 6 of Retransmits for Reno without "bugfix" is illustrated on page 6 of
[F98].) [F98].)
Section 3 of RFC 2582 defined a default variant of NewReno TCP that Section 3 of [RFC2582] defined a default variant of NewReno TCP that
did not use the variable "recover", and did not check if duplicate did not use the variable "recover", and did not check if duplicate
ACKs cover the variable "recover" before invoking Fast Retransmit. ACKs cover the variable "recover" before invoking Fast Retransmit.
With this default variant from RFC 2582, the problem of multiple Fast With this default variant from RFC 2582, the problem of multiple Fast
Retransmits from a single window of data can occur after a Retransmit Retransmits from a single window of data can occur after a Retransmit
Timeout (as in page 8 of [F98]) or in scenarios with reordering (as Timeout (as in page 8 of [F98]) or in scenarios with reordering (as
in the validation test "./test-all-newreno newreno5_noBF" in in the validation test "./test-all-newreno newreno5_noBF" in
directory "tcl/test" of the NS simulator. This gives performance directory "tcl/test" of the NS simulator. This gives performance
similar to that on page 8 of [F03].) RFC 2582 also defined Careful similar to that on page 8 of [F03].) RFC 2582 also defined Careful
and Less Careful variants of the NewReno algorithm, and recommended and Less Careful variants of the NewReno algorithm, and recommended
the Careful variant. the Careful variant.
skipping to change at page 11, line 39 skipping to change at page 10, line 22
If, after a retransmit timeout, the TCP data sender retransmits three If, after a retransmit timeout, the TCP data sender retransmits three
consecutive packets that have already been received by the data consecutive packets that have already been received by the data
receiver, then the TCP data sender will receive three duplicate receiver, then the TCP data sender will receive three duplicate
acknowledgements that do not cover more than "recover". In this acknowledgements that do not cover more than "recover". In this
case, the duplicate acknowledgements are not an indication of a new case, the duplicate acknowledgements are not an indication of a new
instance of congestion. They are simply an indication that the instance of congestion. They are simply an indication that the
sender has unnecessarily retransmitted at least three packets. sender has unnecessarily retransmitted at least three packets.
However, when a retransmitted packet is itself dropped, the sender However, when a retransmitted packet is itself dropped, the sender
can also receive three duplicate acknowledgements that do not cover can also receive three duplicate acknowledgements that do not cover
more than "recover", and in this case the sender would have been more than "recover". In this case, the sender would have been better
better off if it had initiated Fast Retransmit. For a TCP that off if it had initiated Fast Retransmit. For a TCP that implements
implements the algorithm specified in Section 3 of this document, the the algorithm specified in Section 3 of this document, the sender
sender does not infer a packet drop from duplicate acknowledgements does not infer a packet drop from duplicate acknowledgements in this
in this scenario. As always, the retransmit timer is the backup scenario. As always, the retransmit timer is the backup mechanism
mechanism for inferring packet loss in this case. for inferring packet loss in this case.
There are several heuristics, based on timestamps or on the amount of There are several heuristics, based on timestamps or on the amount of
advancement of the cumulative acknowledgement field, that allow the advancement of the cumulative acknowledgement field, that allow the
sender to distinguish in some cases between three duplicate sender to distinguish, in some cases, between three duplicate
acknowledgements following a retransmitted packet that was dropped, acknowledgements following a retransmitted packet that was dropped,
and three duplicate acknowledgements simply from the unnecessary and three duplicate acknowledgements from the unnecessary
retransmission of three packets [Gur03]. The TCP sender MAY use such retransmission of three packets [Gur03, GF04]. The TCP sender MAY
a heuristic to decide to invoke a Fast Retransmit in some cases even use such a heuristic to decide to invoke a Fast Retransmit in some
when the three duplicate acknowledgements do not cover more than cases, even when the three duplicate acknowledgements do not cover
"recover". more than "recover".
For example, when three duplicate acknowledgements are caused by the For example, when three duplicate acknowledgements are caused by the
unnecessary retransmission of three packets, this is likely to be unnecessary retransmission of three packets, this is likely to be
accompanied by the cumulative acknowledgement field advancing by at accompanied by the cumulative acknowledgement field advancing by at
least four segments. Similarly, a heuristic based on timestamps uses least four segments. Similarly, a heuristic based on timestamps uses
the fact that when there is a hole in the sequence space, the the fact that when there is a hole in the sequence space, the
timestamp echoed in the duplicate acknowledgement is the timestamp of timestamp echoed in the duplicate acknowledgement is the timestamp of
the most recent data packet that advanced the cumulative the most recent data packet that advanced the cumulative
acknowledgement field [RFC1323]. If timestamps are used, and the acknowledgement field [RFC1323]. If timestamps are used, and the
sender stores the timestamp of the last acknowledged segment, then sender stores the timestamp of the last acknowledged segment, then
the timestamp echoed by duplicate acknowledgements can be used to the timestamp echoed by duplicate acknowledgements can be used to
distinguish between a retransmitted packet that was dropped, and distinguish between a retransmitted packet that was dropped and three
three duplicate acknowledgements simply from the unnecessary duplicate acknowledgements from the unnecessary retransmission of
retransmission of three packets. The heuristics are illustrated in three packets. The heuristics are illustrated in the NS simulator in
the NS simulator in the validation test "./test-all-newreno". the validation test "./test-all-newreno".
6.1. ACK Heuristic. 6.1. ACK Heuristic
If the ACK-based heuristic is used, then following the advancement of If the ACK-based heuristic is used, then following the advancement of
the cumulative acknowledgement field, the sender stores the value of the cumulative acknowledgement field, the sender stores the value of
previous cumulative acknowledgement as prev_highest_ack and stores the previous cumulative acknowledgement as prev_highest_ack, and
the latest cumulative ACK as highest_ack. In addition, the following stores the latest cumulative ACK as highest_ack. In addition, the
step is performed if Step 1 in Section 3 fails, before proceeding to following step is performed if Step 1 in Section 3 fails, before
Step 1B. proceeding to Step 1B.
1*) If the Cumulative Acknowledgement field didn't cover more than 1*) If the Cumulative Acknowledgement field didn't cover more than
"recover", check to see if the congestion window is greater "recover", check to see if the congestion window is greater than
than SMSS bytes and the difference between highest_ack and SMSS bytes and the difference between highest_ack and
prev_highest_ack is at most 4*SMSS bytes. If true, duplicate prev_highest_ack is at most 4*SMSS bytes. If true, duplicate
ACKs indicate a lost segment (proceed to Step 1A in Section ACKs indicate a lost segment (proceed to Step 1A in Section 3).
3). Otherwise, duplicate ACKs likely result from unnecessary Otherwise, duplicate ACKs likely result from unnecessary
retransmissions (proceed to Step 1B in Section 3). retransmissions (proceed to Step 1B in Section 3).
The congestion window check serves to protect against fast retransmit The congestion window check serves to protect against fast retransmit
immediately after a retransmit timeout, similar to the immediately after a retransmit timeout, similar to the
"exitFastRetrans_" variable in NS. Examples of applying the ACK "exitFastRetrans_" variable in NS. Examples of applying the ACK
heuristic are in validation tests "./test-all-newreno heuristic are in validation tests "./test-all-newreno
newreno_rto_loss_ack" and "./test-all-newreno newreno_rto_dup_ack" in newreno_rto_loss_ack" and "./test-all-newreno newreno_rto_dup_ack" in
directory "tcl/test" of the NS simulator. directory "tcl/test" of the NS simulator.
If several ACKs are lost, the sender can see a jump in the cumulative If several ACKs are lost, the sender can see a jump in the cumulative
ACK of more than three segments and the heuristic can fail. A ACK of more than three segments, and the heuristic can fail. A
validation test for this scenario is "./test-all-newreno validation test for this scenario is "./test-all-newreno
newreno_rto_loss_ackf". RFC 2581 recommends that a receiver should newreno_rto_loss_ackf". RFC 2581 recommends that a receiver should
send duplicate ACKs for every out-of-order data packet, such as a send duplicate ACKs for every out-of-order data packet, such as a
data packet received during Fast Recovery. The ACK heuristic is more data packet received during Fast Recovery. The ACK heuristic is more
likely to fail if the receiver does not follow this advice, because likely to fail if the receiver does not follow this advice, because
then a smaller number of ACK losses are needed to produce a then a smaller number of ACK losses are needed to produce a
sufficient jump in the cumulative ACK. sufficient jump in the cumulative ACK.
6.2. Timestamp Heuristic. 6.2. Timestamp Heuristic
If this heuristic is used, the sender stores the timestamp of the If this heuristic is used, the sender stores the timestamp of the
last acknowledged segment. In addition, the second paragraph of step last acknowledged segment. In addition, the second paragraph of step
1 in Section 3 is replaced as follows: 1 in Section 3 is replaced as follows:
1**) If the Cumulative Acknowledgement field didn't cover more than 1**) If the Cumulative Acknowledgement field didn't cover more than
"recover", check to see if the echoed timestamp equals the "recover", check to see if the echoed timestamp in the last
stored timestamp. If true, duplicate ACKs indicate a lost non-duplicate acknowledgment equals the stored timestamp. If
segment (proceed to Step 1A in Section 3). Otherwise, duplicate true, duplicate ACKs indicate a lost segment (proceed to Step 1A
ACKs likely result from unnecessary retransmissions (proceed in Section 3). Otherwise, duplicate ACKs likely result from
to Step 1B in Section 3). unnecessary retransmissions (proceed to Step 1B in Section 3).
Examples of applying the timestamp heuristic are in validation tests Examples of applying the timestamp heuristic are in validation tests
"./test-all-newreno newreno_rto_loss_tsh" and "./test-all-newreno "./test-all-newreno newreno_rto_loss_tsh" and "./test-all-newreno
newreno_rto_dup_tsh". The timestamp heuristic works correctly both newreno_rto_dup_tsh". The timestamp heuristic works correctly, both
when the receiver echoes timestamps as specified by [RFC1323] or by when the receiver echoes timestamps as specified by [RFC1323], and by
its revision attempts. However, if the receiver arbitrarily echos its revision attempts. However, if the receiver arbitrarily echoes
timestamps, the heuristic can fail. The heuristic can also fail if a timestamps, the heuristic can fail. The heuristic can also fail if a
timeout was spurious and returning ACKs are not from retransmitted timeout was spurious and returning ACKs are not from retransmitted
segments. This can be prevented by detection algorithms such as segments. This can be prevented by detection algorithms such as
[RFC3522]. [RFC3522].
7. Implementation Issues for the Data Receiver 7. Implementation Issues for the Data Receiver
[RFC2581] specifies that "Out-of-order data segments SHOULD be [RFC2581] specifies that "Out-of-order data segments SHOULD be
acknowledged immediately, in order to accelerate loss recovery." acknowledged immediately, in order to accelerate loss recovery."
Neal Cardwell has noted that some data receivers do not send an Neal Cardwell has noted that some data receivers do not send an
immediate acknowledgement when they send a partial acknowledgment, immediate acknowledgement when they send a partial acknowledgment,
but instead wait first for their delayed acknowledgement timer to but instead wait first for their delayed acknowledgement timer to
expire [C98]. As [C98] notes, this severely limits the potential expire [C98]. As [C98] notes, this severely limits the potential
benefit from NewReno by delaying the receipt of the partial benefit of NewReno by delaying the receipt of the partial
acknowledgement at the data sender. Echoing RFC 2581, our acknowledgement at the data sender. Echoing RFC 2581, our
recommendation is that the data receiver send an immediate recommendation is that the data receiver send an immediate
acknowledgement for an out-of-order segment, even when that out-of- acknowledgement for an out-of-order segment, even when that out-of-
order segment fills a hole in the buffer. order segment fills a hole in the buffer.
8. Implementation Issues for the Data Sender 8. Implementation Issues for the Data Sender
In Section 3, Step 5 above, it is noted that implementations should In Section 3, Step 5 above, it is noted that implementations should
take measures to avoid a possible burst of data when leaving Fast take measures to avoid a possible burst of data when leaving Fast
Recovery, in case the amount of new data that the sender is eligible Recovery, in case the amount of new data that the sender is eligible
skipping to change at page 14, line 21 skipping to change at page 12, line 52
to a single acknowledgment. (This is known as "maxburst_" in the ns to a single acknowledgment. (This is known as "maxburst_" in the ns
simulator.) Other possible mechanisms for avoiding bursts include simulator.) Other possible mechanisms for avoiding bursts include
rate-based pacing, or setting the slow-start threshold to the rate-based pacing, or setting the slow-start threshold to the
resultant congestion window and then resetting the congestion window resultant congestion window and then resetting the congestion window
to FlightSize. A recommendation on the general mechanism to avoid to FlightSize. A recommendation on the general mechanism to avoid
excessively bursty sending patterns is outside the scope of this excessively bursty sending patterns is outside the scope of this
document. document.
An implementation may want to use a separate flag to record whether An implementation may want to use a separate flag to record whether
or not it is presently in the Fast Recovery procedure. The use of or not it is presently in the Fast Recovery procedure. The use of
the value of the duplicate acknowledgment counter as such a flag is the value of the duplicate acknowledgment counter for this purpose is
not reliable because it can be reset upon window updates and out-of- not reliable because it can be reset upon window updates and out-of-
order acknowledgments. order acknowledgments.
When not in Fast Recovery, the value of the state variable "recover" When not in Fast Recovery, the value of the state variable "recover"
should be pulled along with the value of the state variable for should be pulled along with the value of the state variable for
acknowledgments (typically, "snd_una") so that, when large amounts of acknowledgments (typically, "snd_una") so that, when large amounts of
data has been sent and acked, the sequence space does not wrap and data have been sent and acked, the sequence space does not wrap and
falsely indicate that Fast Recovery should not be entered (Section 3, falsely indicate that Fast Recovery should not be entered (Section 3,
step 1, last paragraph). step 1, last paragraph).
It is important for the sender to respond correctly to duplicate ACKs It is important for the sender to respond correctly to duplicate ACKs
received when the sender is no longer in Fast Recovery (e.g., because received when the sender is no longer in Fast Recovery (e.g., because
of a Retransmit Timeout). The Limited Transmit procedure [RFC3042] of a Retransmit Timeout). The Limited Transmit procedure [RFC3042]
describes possible responses to the first and second duplicate describes possible responses to the first and second duplicate
acknowledgements. When three or more duplicate acknowledgements are acknowledgements. When three or more duplicate acknowledgements are
received, the Cumulative Acknowledgement field doesn't cover more received, the Cumulative Acknowledgement field doesn't cover more
than "recover", and a new Fast Recovery is not invoked, it is than "recover", and a new Fast Recovery is not invoked, it is
skipping to change at page 15, line 21 skipping to change at page 13, line 51
modification described in this document improves the performance of modification described in this document improves the performance of
the Fast Retransmit and Fast Recovery algorithms of Reno TCP in a the Fast Retransmit and Fast Recovery algorithms of Reno TCP in a
wide variety of scenarios. This has been discussed in some depth in wide variety of scenarios. This has been discussed in some depth in
[FF96], which illustrates Reno TCP's poor performance when multiple [FF96], which illustrates Reno TCP's poor performance when multiple
packets are dropped from a window of data and also illustrates packets are dropped from a window of data and also illustrates
NewReno TCP's good performance in that scenario. NewReno TCP's good performance in that scenario.
We do, however, know of one scenario where Reno TCP gives better We do, however, know of one scenario where Reno TCP gives better
performance than NewReno TCP, that we describe here for the sake of performance than NewReno TCP, that we describe here for the sake of
completeness. Consider a scenario with no packet loss, but with completeness. Consider a scenario with no packet loss, but with
sufficient reordering that the TCP sender receives three duplicate sufficient reordering so that the TCP sender receives three duplicate
acknowledgements. This will trigger the Fast Retransmit and Fast acknowledgements. This will trigger the Fast Retransmit and Fast
Recovery algorithms. With Reno TCP or with Sack TCP, this will Recovery algorithms. With Reno TCP or with Sack TCP, this will
result in the unnecessary retransmission of a single packet, combined result in the unnecessary retransmission of a single packet, combined
with a halving of the congestion window (shown on pages 4 and 6 of with a halving of the congestion window (shown on pages 4 and 6 of
[F03]). With NewReno TCP, however, this reordering will also result [F03]). With NewReno TCP, however, this reordering will also result
in the unnecessary retransmission of an entire window of data (shown in the unnecessary retransmission of an entire window of data (shown
on page 5 of [F03]). on page 5 of [F03]).
While Reno TCP performs better than NewReno TCP in the presence of While Reno TCP performs better than NewReno TCP in the presence of
reordering, NewReno's superior performance in the presence of reordering, NewReno's superior performance in the presence of
skipping to change at page 15, line 49 skipping to change at page 14, line 31
implementations in the Internet today, as documented in [PF01]. For implementations in the Internet today, as documented in [PF01]. For
example, tests of TCP implementations in several thousand web servers example, tests of TCP implementations in several thousand web servers
in 2001 showed that for those TCP connections where the web browser in 2001 showed that for those TCP connections where the web browser
was not SACK-capable, more web servers used the Fast Retransmit and was not SACK-capable, more web servers used the Fast Retransmit and
Fast Recovery algorithms of NewReno than those of Reno or Tahoe TCP Fast Recovery algorithms of NewReno than those of Reno or Tahoe TCP
[PF01]. [PF01].
11. Changes Relative to RFC 2582 11. Changes Relative to RFC 2582
The purpose of this document is to advance the NewReno's Fast The purpose of this document is to advance the NewReno's Fast
Retransmit and Fast Recovery algorithms in RFC 2582 to Proposed Retransmit and Fast Recovery algorithms in RFC 2582 to Standards
Standard. Track.
The main change in this document relative to RFC 2582 is to specify The main change in this document relative to RFC 2582 is to specify
the Careful variant of NewReno's Fast Retransmit and Fast Recovery the Careful variant of NewReno's Fast Retransmit and Fast Recovery
algorithms. The base algorithm described in RFC 2582 did not attempt algorithms. The base algorithm described in RFC 2582 did not attempt
to avoid unnecessary multiple Fast Retransmits that can occur after a to avoid unnecessary multiple Fast Retransmits that can occur after a
timeout (described in more detail in the section above). However, timeout (described in more detail in the section above). However,
RFC 2582 also defined "Careful" and "Less Careful" variants that RFC 2582 also defined "Careful" and "Less Careful" variants that
avoid these unnecessary Fast Retransmits, and recommended the Careful avoid these unnecessary Fast Retransmits, and recommended the Careful
variant. This document specifies the previously-named "Careful" variant. This document specifies the previously-named "Careful"
variant as the basic version of NewReno. As described below, this variant as the basic version of NewReno. As described below, this
algorithm uses a variable "recover", whose initial value is the send algorithm uses a variable "recover", whose initial value is the send
sequence number. sequence number.
The algorithm specified in Section 3 checks whether the The algorithm specified in Section 3 checks whether the
acknowledgement field of a partial acknowledgement covers *more* than acknowledgement field of a partial acknowledgement covers *more* than
"recover". Another possible variant would be to require simply that "recover", as defined in Section 3. Another possible variant would
the acknowledgement field *cover* "recover" before initiating another be to simply require that the acknowledgement field covers *more than
Fast Retransmit. We called this the Less Careful variant in RFC or equal to* "recover" before initiating another Fast Retransmit. We
2582. called this the Less Careful variant in RFC 2582.
There are two separate scenarios in which the TCP sender could There are two separate scenarios in which the TCP sender could
receive three duplicate acknowledgements acknowledging "recover" but receive three duplicate acknowledgements acknowledging "recover" but
no more than "recover". One scenario would be that the data sender no more than "recover". One scenario would be that the data sender
transmitted four packets with sequence numbers higher than "recover", transmitted four packets with sequence numbers higher than "recover",
that the first packet was dropped in the network, and the following that the first packet was dropped in the network, and the following
three packets triggered three duplicate acknowledgements three packets triggered three duplicate acknowledgements
acknowledging "recover". The second scenario would be that the acknowledging "recover". The second scenario would be that the
sender unnecessarily retransmitted three packets below "recover", and sender unnecessarily retransmitted three packets below "recover", and
that these three packets triggered three duplicate acknowledgements that these three packets triggered three duplicate acknowledgements
acknowledging "recover". In the absence of SACK, the TCP sender in acknowledging "recover". In the absence of SACK, the TCP sender is
unable to distinguish between these two scenarios. unable to distinguish between these two scenarios.
For the Careful variant of Fast Retransmit, the data sender would For the Careful variant of Fast Retransmit, the data sender would
have to wait for a retransmit timeout in the first scenario, but have to wait for a retransmit timeout in the first scenario, but
would not have an unnecessary Fast Retransmit in the second scenario. would not have an unnecessary Fast Retransmit in the second scenario.
For the Less Careful variant to Fast Retransmit, the data sender For the Less Careful variant to Fast Retransmit, the data sender
would Fast Retransmit as desired in the first scenario, and would would Fast Retransmit as desired in the first scenario, and would
unnecessarily Fast Retransmit in the second scenario. This document unnecessarily Fast Retransmit in the second scenario. This document
only specifies the Careful variant in Section 3. Unnecessary Fast only specifies the Careful variant in Section 3. Unnecessary Fast
Retransmits with the Less Careful variant in scenarios with Retransmits with the Less Careful variant in scenarios with
skipping to change at page 17, line 8 skipping to change at page 15, line 36
The document also specifies two heuristics that the TCP sender MAY The document also specifies two heuristics that the TCP sender MAY
use to decide to invoke Fast Retransmit even when the three duplicate use to decide to invoke Fast Retransmit even when the three duplicate
acknowledgements do not cover more than "recover". These heuristics, acknowledgements do not cover more than "recover". These heuristics,
an ACK-based heuristic and a timestamp heuristic, are described in an ACK-based heuristic and a timestamp heuristic, are described in
Sections 6.1 and 6.2 respectively. Sections 6.1 and 6.2 respectively.
12. Conclusions 12. Conclusions
This document specifies the NewReno Fast Retransmit and Fast Recovery This document specifies the NewReno Fast Retransmit and Fast Recovery
algorithms for TCP. This NewReno modification to TCP can be algorithms for TCP. This NewReno modification to TCP can even be
important even for TCP implementations that support the SACK option, important for TCP implementations that support the SACK option,
because the SACK option can only be used for TCP connections when because the SACK option can only be used for TCP connections when
both TCP end-nodes support the SACK option. NewReno performs better both TCP end-nodes support the SACK option. NewReno performs better
than Reno (RFC 2581) in a number of scenarios discussed herein. than Reno (RFC 2581) in a number of scenarios discussed herein.
A number of options to the basic algorithm presented in Section 3 are A number of options to the basic algorithm presented in Section 3 are
also described. These include the handling of the retransmission also described. These include the handling of the retransmission
timer (Section 4), the response to partial acknowledgments (Section timer (Section 4), the response to partial acknowledgments (Section
5), and the value of the congestion window when leaving Fast Recovery 5), and the value of the congestion window when leaving Fast Recovery
(section 3, step 5). Our belief is that the differences between (section 3, step 5). Our belief is that the differences between
these variants of NewReno are small compared to the differences these variants of NewReno are small compared to the differences
between Reno and NewReno. That is, the important thing is to between Reno and NewReno. That is, the important thing is to
implement NewReno instead of Reno, for a TCP connection without SACK; implement NewReno instead of Reno, for a TCP connection without SACK;
it is less important exactly which of the variants of NewReno is it is less important exactly which of the variants of NewReno is
implemented. implemented.
13. Acknowledgements 13. Security Considerations
RFC 2581 discusses general security considerations concerning TCP
congestion control. This document describes a specific algorithm
that conforms with the congestion control requirements of RFC 2581,
and so those considerations apply to this algorithm, too. There are
no known additional security concerns for this specific algorithm.
14. Acknowledgements
Many thanks to Anil Agarwal, Mark Allman, Armando Caro, Jeffrey Hsu, Many thanks to Anil Agarwal, Mark Allman, Armando Caro, Jeffrey Hsu,
Vern Paxson, Kacheong Poon, Keyur Shah, and Bernie Volz for detailed Vern Paxson, Kacheong Poon, Keyur Shah, and Bernie Volz for detailed
feedback on this document or on its precursor RFC 2582. feedback on this document or on its precursor, RFC 2582.
14. References 15. References
Normative References 15.1. Normative References
[RFC2018] M. Mathis, J. Mahdavi, S. Floyd, A. Romanow, "TCP Selective [RFC2018] Mathis, M., Mahdavi, J., Floyd, S. and A. Romanow, "TCP
Acknowledgement Options", RFC 2018, October 1996. Selective Acknowledgement Options", RFC 2018, October 1996.
[RFC2581] W. Stevens, M. Allman, and V. Paxson, "TCP Congestion [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate
Requirement Levels", BCP 14, RFC 2119, March 1997.
[RFC2581] Allman, M., Paxson, V. and W. Stevens, "TCP Congestion
Control", RFC 2581, April 1999. Control", RFC 2581, April 1999.
[RFC2582] S. Floyd and T. Henderson, The NewReno Modification to [RFC2582] Floyd, S. and T. Henderson, "The NewReno Modification to
TCP's Fast Recovery Algorithm, RFC 2582, April 1999. TCP's Fast Recovery Algorithm", RFC 2582, April 1999.
[RFC2988] V. Paxson and M. Allman, Computing TCP's Retransmission [RFC2988] Paxson, V. and M. Allman, "Computing TCP's Retransmission
Timer, RFC 2988, November 2000. Timer", RFC 2988, November 2000.
[RFC3042] M. Allman, H. Balakrishnan, and S. Floyd, Enhancing TCP's [RFC3042] Allman, M., Balakrishnan, H. and S. Floyd, "Enhancing TCP's
Loss Recovery Using Limited Transmit, RFC 3042, January 2001. Loss Recovery Using Limited Transmit", RFC 3042, January
2001.
Informative References 15.2. Informative References
[C98] N. Cardwell, "delayed ACKs for retransmitted packets: ouch!". [C98] Cardwell, N., "delayed ACKs for retransmitted packets:
November 1998, Email to the tcpimpl mailing list, Message-ID ouch!". November 1998, Email to the tcpimpl mailing list,
"Pine.LNX.4.02A.9811021421340.26785-100000@sake.cs.washington.edu", Message-ID "Pine.LNX.4.02A.9811021421340.26785-
archived at "http://tcp-impl.lerc.nasa.gov/tcp-impl". 100000@sake.cs.washington.edu", archived at "http://tcp-
impl.lerc.nasa.gov/tcp-impl".
[F98] S. Floyd, Revisions to RFC 2001, "Presentation to the TCPIMPL [F98] Floyd, S., Revisions to RFC 2001, "Presentation to the
Working Group", August 1998. URLs "ftp://ftp.ee.lbl.gov/talks/sf- TCPIMPL Working Group", August 1998. URLs
tcpimpl-aug98.ps" and "ftp://ftp.ee.lbl.gov/talks/sf-tcpimpl- "ftp://ftp.ee.lbl.gov/talks/sf-tcpimpl-aug98.ps" and
aug98.pdf". "ftp://ftp.ee.lbl.gov/talks/sf-tcpimpl-aug98.pdf".
[F03] S. Floyd, "Moving NewReno from Experimental to Proposed [F03] Floyd, S., "Moving NewReno from Experimental to Proposed
Standard? Presentation to the TSVWG Working Group", March 2003. Standard? Presentation to the TSVWG Working Group", March
URLs "http://www.icir.org/floyd/talks/newreno-Mar03.ps" and 2003. URLs "http://www.icir.org/floyd/talks/newreno-
"http://www.icir.org/floyd/talks/newreno-Mar03.pdf". Mar03.ps" and "http://www.icir.org/floyd/talks/newreno-
Mar03.pdf".
[FF96] K. Fall and S. Floyd, "Simulation-based Comparisons of Tahoe, [FF96] Fall, K. and S. Floyd, "Simulation-based Comparisons of
Reno and SACK TCP", Computer Communication Review, July 1996. URL Tahoe, Reno and SACK TCP", Computer Communication Review,
"ftp://ftp.ee.lbl.gov/papers/sacks.ps.Z". July 1996. URL "ftp://ftp.ee.lbl.gov/papers/sacks.ps.Z".
[F94] S. Floyd, "TCP and Successive Fast Retransmits", Technical [F94] Floyd, S., "TCP and Successive Fast Retransmits", Technical
report, October 1994. URL report, October 1994. URL
"ftp://ftp.ee.lbl.gov/papers/fastretrans.ps". "ftp://ftp.ee.lbl.gov/papers/fastretrans.ps".
[Gur03] A. Gurtov, "[Tsvwg] resolving the problem of unnecessary fast [GF04] Gurtov, A. and S. Floyd, "Resolving Acknowledgment
retransmits in go-back-N", email to the tsvwg mailing list, message Ambiguity in non-SACK TCP", Next Generation Teletraffic and
ID <3F25B467.9020609@cs.helsinki.fi>, July 28, 2003. URL Wired/Wireless Advanced Networking (NEW2AN'04), February
"http://www1.ietf.org/mail-archive/working- 2004. URL "http://www.cs.helsinki.fi/u/gurtov/papers/
heuristics.html".
[Gur03] Gurtov, A., "[Tsvwg] resolving the problem of unnecessary
fast retransmits in go-back-N", email to the tsvwg mailing
list, message ID <3F25B467.9020609@cs.helsinki.fi>, July
28, 2003. URL "http://www1.ietf.org/mail-archive/working-
groups/tsvwg/current/msg04334.html". groups/tsvwg/current/msg04334.html".
[Hen98] T. Henderson, Re: NewReno and the 2001 Revision. September [Hen98] Henderson, T., Re: NewReno and the 2001 Revision. September
1998. Email to the tcpimpl mailing list, Message ID 1998. Email to the tcpimpl mailing list, Message ID
"Pine.BSI.3.95.980923224136.26134A-100000@raptor.CS.Berkeley.EDU", "Pine.BSI.3.95.980923224136.26134A-
archived at "http://tcp-impl.lerc.nasa.gov/tcp-impl". 100000@raptor.CS.Berkeley.EDU", archived at "http://tcp-
impl.lerc.nasa.gov/tcp-impl".
[Hoe95] J. Hoe, "Startup Dynamics of TCP's Congestion Control and [Hoe95] Hoe, J., "Startup Dynamics of TCP's Congestion Control and
Avoidance Schemes", Master's Thesis, MIT, 1995. URL "http://ana- Avoidance Schemes", Master's Thesis, MIT, 1995.
www.lcs.mit.edu/anaweb/ps-papers/hoe-thesis.ps".
[Hoe96] J. Hoe, "Improving the Start-up Behavior of a Congestion [Hoe96] Hoe, J., "Improving the Start-up Behavior of a Congestion
Control Scheme for TCP", ACM SIGCOMM, August 1996. URL Control Scheme for TCP", ACM SIGCOMM, August 1996. URL
"http://www.acm.org/sigcomm/sigcomm96/program.html". "http://www.acm.org/sigcomm/sigcomm96/program.html".
[LM97] D. Lin and R. Morris, "Dynamics of Random Early Detection", [LM97] Lin, D. and R. Morris, "Dynamics of Random Early
SIGCOMM 97, September 1997. URL Detection", SIGCOMM 97, September 1997. URL
"http://www.acm.org/sigcomm/sigcomm97/program.html". "http://www.acm.org/sigcomm/sigcomm97/program.html".
[NS] The Network Simulator (NS). URL "http://www.isi.edu/nsnam/ns/". [NS] The Network Simulator (NS). URL
"http://www.isi.edu/nsnam/ns/".
[PF01] J. Padhye and S. Floyd, "Identifying the TCP Behavior of Web
Servers", June 2001, SIGCOMM 2001.
[RFC1323] V. Jacobson, R. Braden, and D. Borman, "TCP Extensions for
High Performance,", RFC 1323, May 1992.
[RFC3517] E. Blanton, M. Allman, K. Fall, and L. Wang, "A [PF01] Padhye, J. and S. Floyd, "Identifying the TCP Behavior of
Conservative Selective Acknowledgment (SACK)-based Loss Recovery Web Servers", June 2001, SIGCOMM 2001.
Algorithm for TCP", RFC 3517, April 2003.
[RFC3522] R. Ludwig and M. Meyer, The Eifel Detection Algorithm for [RFC1323] Jacobson, V., Braden, R. and D. Borman, "TCP Extensions for
TCP, RFC 3522, April 2003. High Performance", RFC 1323, May 1992.
15. Security Considerations [RFC3517] Blanton, E., Allman, M., Fall, K. and L. Wang, "A
Conservative Selective Acknowledgment (SACK)-based Loss
Recovery Algorithm for TCP", RFC 3517, April 2003.
RFC 2581 discusses general security considerations concerning TCP [RFC3522] Ludwig, R. and M. Meyer, "The Eifel Detection Algorithm for
congestion control. This document describes a specific algorithm TCP", RFC 3522, April 2003.
that conforms with the congestion control requirements of RFC 2581,
and so those considerations apply to this algorithm, too. There are
no known additional security concerns for this specific algorithm.
AUTHORS' ADDRESSES Authors' Addresses
Sally Floyd Sally Floyd
International Computer Science Institute International Computer Science Institute
Phone: +1 (510) 666-2989 Phone: +1 (510) 666-2989
Email: floyd@acm.org EMail: floyd@acm.org
URL: http://www.icir.org/floyd/ URL: http://www.icir.org/floyd/
Tom Henderson Tom Henderson
The Boeing Company The Boeing Company
Email: thomas.r.henderson@boeing.com EMail: thomas.r.henderson@boeing.com
Andrei Gurtov Andrei Gurtov
U. Helsinki TeliaSonera
Email: gurtov@cs.helsinki.fi EMail: andrei.gurtov@teliasonera.com
Full Copyright Statement
Copyright (C) The Internet Society (2004). This document is subject
to the rights, licenses and restrictions contained in BCP 78, and
except as set forth therein, the authors retain all their rights.
This document and the information contained herein are provided on an
"AS IS" basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE
REPRESENTS OR IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY AND THE
INTERNET ENGINEERING TASK FORCE DISCLAIM ALL WARRANTIES, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF
THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED
WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.
Intellectual Property
The IETF takes no position regarding the validity or scope of any
Intellectual Property Rights or other rights that might be claimed
to pertain to the implementation or use of the technology
described in this document or the extent to which any license
under such rights might or might not be available; nor does it
represent that it has made any independent effort to identify any
such rights. Information on the procedures with respect to
rights in RFC documents can be found in BCP 78 and BCP 79.
Copies of IPR disclosures made to the IETF Secretariat and any
assurances of licenses to be made available, or the result of an
attempt made to obtain a general license or permission for the use
of such proprietary rights by implementers or users of this
specification can be obtained from the IETF on-line IPR repository
at http://www.ietf.org/ipr.
The IETF invites any interested party to bring to its attention
any copyrights, patents or patent applications, or other
proprietary rights that may cover technology that may be required
to implement this standard. Please address the information to the
IETF at ietf-ipr@ietf.org.
Acknowledgement
Funding for the RFC Editor function is currently provided by the
Internet Society.
 End of changes. 

This html diff was produced by rfcdiff 1.25, available from http://www.levkowetz.com/ietf/tools/rfcdiff/