draft-ietf-tsvwg-byte-pkt-congest-01.txt   draft-ietf-tsvwg-byte-pkt-congest-02.txt 
Transport Area Working Group B. Briscoe Transport Area Working Group B. Briscoe
Internet-Draft BT Internet-Draft BT
Updates: 2309 (if approved) October 23, 2009 Updates: 2309 (if approved) J. Manner
Intended status: Informational Intended status: Informational Aalto University
Expires: April 26, 2010 Expires: January 13, 2011 July 12, 2010
Byte and Packet Congestion Notification Byte and Packet Congestion Notification
draft-ietf-tsvwg-byte-pkt-congest-01 draft-ietf-tsvwg-byte-pkt-congest-02
Status of this Memo Abstract
This Internet-Draft is submitted to IETF in full conformance with the This memo concerns dropping or marking packets using active queue
management (AQM) such as random early detection (RED) or pre-
congestion notification (PCN). We give two strong recommendations:
(1) packet size should not be taken into account when transports read
congestion indications, not when network equipment writes them, and
(2) byte-mode packet drop variant of AQM algorithms, such as RED,
should not be used to drop fewer small packets.
Status of This Memo
This Internet-Draft is submitted in full conformance with the
provisions of BCP 78 and BCP 79. provisions of BCP 78 and BCP 79.
Internet-Drafts are working documents of the Internet Engineering Internet-Drafts are working documents of the Internet Engineering
Task Force (IETF), its areas, and its working groups. Note that Task Force (IETF). Note that other groups may also distribute
other groups may also distribute working documents as Internet- working documents as Internet-Drafts. The list of current Internet-
Drafts. Drafts is at http://datatracker.ietf.org/drafts/current/.
Internet-Drafts are draft documents valid for a maximum of six months Internet-Drafts are draft documents valid for a maximum of six months
and may be updated, replaced, or obsoleted by other documents at any and may be updated, replaced, or obsoleted by other documents at any
time. It is inappropriate to use Internet-Drafts as reference time. It is inappropriate to use Internet-Drafts as reference
material or to cite them other than as "work in progress." material or to cite them other than as "work in progress."
The list of current Internet-Drafts can be accessed at This Internet-Draft will expire on January 13, 2011.
http://www.ietf.org/ietf/1id-abstracts.txt.
The list of Internet-Draft Shadow Directories can be accessed at
http://www.ietf.org/shadow.html.
This Internet-Draft will expire on April 26, 2010.
Copyright Notice Copyright Notice
Copyright (c) 2009 IETF Trust and the persons identified as the Copyright (c) 2010 IETF Trust and the persons identified as the
document authors. All rights reserved. document authors. All rights reserved.
This document is subject to BCP 78 and the IETF Trust's Legal This document is subject to BCP 78 and the IETF Trust's Legal
Provisions Relating to IETF Documents in effect on the date of Provisions Relating to IETF Documents
publication of this document (http://trustee.ietf.org/license-info). (http://trustee.ietf.org/license-info) in effect on the date of
Please review these documents carefully, as they describe your rights publication of this document. Please review these documents
and restrictions with respect to this document. carefully, as they describe your rights and restrictions with respect
to this document. Code Components extracted from this document must
Abstract include Simplified BSD License text as described in Section 4.e of
the Trust Legal Provisions and are provided without warranty as
This memo concerns dropping or marking packets using active queue described in the Simplified BSD License.
management (AQM) such as random early detection (RED) or pre-
congestion notification (PCN). The primary conclusion is that packet
size should be taken into account when transports read congestion
indications, not when network equipment writes them. Reducing drop
of small packets has some tempting advantages: i) it drops less
control packets, which tend to be small and ii) it makes TCP's bit-
rate less dependent on packet size. However, there are ways of
addressing these issues at the transport layer, rather than reverse
engineering network forwarding to fix specific transport problems.
Network layer algorithms like the byte-mode packet drop variant of
RED should not be used to drop fewer small packets, because that
creates a perverse incentive for transports to use tiny segments,
consequently also opening up a DoS vulnerability.
Table of Contents Table of Contents
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 6 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.1. Requirements Notation . . . . . . . . . . . . . . . . . . 9 1.1. Terminology and Scoping . . . . . . . . . . . . . . . . . 6
2. Motivating Arguments . . . . . . . . . . . . . . . . . . . . . 9 1.2. Why now? . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.1. Scaling Congestion Control with Packet Size . . . . . . . 9 2. Motivating Arguments . . . . . . . . . . . . . . . . . . . . . 8
2.1. Scaling Congestion Control with Packet Size . . . . . . . 8
2.2. Avoiding Perverse Incentives to (ab)use Smaller Packets . 10 2.2. Avoiding Perverse Incentives to (ab)use Smaller Packets . 10
2.3. Small != Control . . . . . . . . . . . . . . . . . . . . . 12 2.3. Small != Control . . . . . . . . . . . . . . . . . . . . . 11
2.4. Implementation Efficiency . . . . . . . . . . . . . . . . 12 2.4. Implementation Efficiency . . . . . . . . . . . . . . . . 11
3. Working Definition of Congestion Notification . . . . . . . . 12 3. The State of the Art . . . . . . . . . . . . . . . . . . . . . 11
4. Congestion Measurement . . . . . . . . . . . . . . . . . . . . 13 3.1. Congestion Measurement: Status . . . . . . . . . . . . . . 12
4.1. Congestion Measurement by Queue Length . . . . . . . . . . 13 3.1.1. Fixed Size Packet Buffers . . . . . . . . . . . . . . 13
4.1.1. Fixed Size Packet Buffers . . . . . . . . . . . . . . 13 3.1.2. Congestion Measurement without a Queue . . . . . . . . 14
4.2. Congestion Measurement without a Queue . . . . . . . . . . 14 3.2. Congestion Coding: Status . . . . . . . . . . . . . . . . 14
5. Idealised Wire Protocol Coding . . . . . . . . . . . . . . . . 15 3.2.1. Network Bias when Encoding . . . . . . . . . . . . . . 14
6. The State of the Art . . . . . . . . . . . . . . . . . . . . . 17 3.2.2. Transport Bias when Decoding . . . . . . . . . . . . . 16
6.1. Congestion Measurement: Status . . . . . . . . . . . . . . 17 3.2.3. Making Transports Robust against Control Packet
6.2. Congestion Coding: Status . . . . . . . . . . . . . . . . 18 Losses . . . . . . . . . . . . . . . . . . . . . . . . 17
6.2.1. Network Bias when Encoding . . . . . . . . . . . . . . 18 3.2.4. Congestion Coding: Summary of Status . . . . . . . . . 18
6.2.2. Transport Bias when Decoding . . . . . . . . . . . . . 20 4. Outstanding Issues and Next Steps . . . . . . . . . . . . . . 20
6.2.3. Making Transports Robust against Control Packet 4.1. Bit-congestible World . . . . . . . . . . . . . . . . . . 20
Losses . . . . . . . . . . . . . . . . . . . . . . . . 21 4.2. Bit- & Packet-congestible World . . . . . . . . . . . . . 21
6.2.4. Congestion Coding: Summary of Status . . . . . . . . . 22 5. Recommendation and Conclusions . . . . . . . . . . . . . . . . 22
7. Outstanding Issues and Next Steps . . . . . . . . . . . . . . 24 5.1. Recommendation on Queue Measurement . . . . . . . . . . . 22
7.1. Bit-congestible World . . . . . . . . . . . . . . . . . . 24 5.2. Recommendation on Notifying Congestion . . . . . . . . . . 23
7.2. Bit- & Packet-congestible World . . . . . . . . . . . . . 24 5.3. Recommendation on Responding to Congestion . . . . . . . . 24
8. Security Considerations . . . . . . . . . . . . . . . . . . . 25 5.4. Recommended Future Research . . . . . . . . . . . . . . . 24
9. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . 26 6. Security Considerations . . . . . . . . . . . . . . . . . . . 24
10. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . 28 7. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . 25
11. Comments Solicited . . . . . . . . . . . . . . . . . . . . . . 28 8. Comments Solicited . . . . . . . . . . . . . . . . . . . . . . 25
12. References . . . . . . . . . . . . . . . . . . . . . . . . . . 28 9. References . . . . . . . . . . . . . . . . . . . . . . . . . . 25
12.1. Normative References . . . . . . . . . . . . . . . . . . . 28 9.1. Normative References . . . . . . . . . . . . . . . . . . . 25
12.2. Informative References . . . . . . . . . . . . . . . . . . 29 9.2. Informative References . . . . . . . . . . . . . . . . . . 26
Editorial Comments . . . . . . . . . . . . . . . . . . . . . . . . Appendix A. Congestion Notification Definition: Further
Appendix A. Example Scenarios . . . . . . . . . . . . . . . . . . 32 Justification . . . . . . . . . . . . . . . . . . . . 30
A.1. Notation . . . . . . . . . . . . . . . . . . . . . . . . . 32 Appendix B. Idealised Wire Protocol . . . . . . . . . . . . . . . 30
A.2. Bit-congestible resource, equal bit rates (Ai) . . . . . . 32 B.1. Protocol Coding . . . . . . . . . . . . . . . . . . . . . 30
A.3. Bit-congestible resource, equal packet rates (Bi) . . . . 33 B.2. Example Scenarios . . . . . . . . . . . . . . . . . . . . 32
A.4. Pkt-congestible resource, equal bit rates (Aii) . . . . . 34 B.2.1. Notation . . . . . . . . . . . . . . . . . . . . . . . 32
A.5. Pkt-congestible resource, equal packet rates (Bii) . . . . 35 B.2.2. Bit-congestible resource, equal bit rates (Ai) . . . . 32
Appendix B. Congestion Notification Definition: Further B.2.3. Bit-congestible resource, equal packet rates (Bi) . . 33
Justification . . . . . . . . . . . . . . . . . . . . 35 B.2.4. Pkt-congestible resource, equal bit rates (Aii) . . . 34
Appendix C. Byte-mode Drop Complicates Policing Congestion B.2.5. Pkt-congestible resource, equal packet rates (Bii) . . 35
Response . . . . . . . . . . . . . . . . . . . . . . 36
Author's Address . . . . . . . . . . . . . . . . . . . . . . . . . 37
Changes from Previous Versions
To be removed by the RFC Editor on publication.
Full incremental diffs between each version are available at
<http://www.cs.ucl.ac.uk/staff/B.Briscoe/pubs.html#byte-pkt-congest>
or
<http://tools.ietf.org/wg/tsvwg/draft-ietf-tsvwg-byte-pkt-congest/>
(courtesy of the rfcdiff tool):
From -00 to -01 (this version):
* Minor clarifications throughout and updated references
From briscoe-byte-pkt-mark-02 to ietf-byte-pkt-congest-00:
* Added note on relationship to existing RFCs
* Posed the question of whether packet-congestion could become
common and deferred it to the IRTF ICCRG. Added ref to the
dual-resource queue (DRQ) proposal.
* Changed PCN references from the PCN charter & architecture to
the PCN marking behaviour draft most likely to imminently
become the standards track WG item.
From -01 to -02:
* Abstract reorganised to align with clearer separation of issue
in the memo.
* Introduction reorganised with motivating arguments removed to
new Section 2.
* Clarified avoiding lock-out of large packets is not the main or
only motivation for RED.
* Mentioned choice of drop or marking explicitly throughout,
rather than trying to coin a word to mean either.
* Generalised the discussion throughout to any packet forwarding
function on any network equipment, not just routers.
* Clarified the last point about why this is a good time to sort
out this issue: because it will be hard / impossible to design
new transports unless we decide whether the network or the
transport is allowing for packet size.
* Added statement explaining the horizon of the memo is long
term, but with short term expediency in mind.
* Added material on scaling congestion control with packet size
(Section 2.1).
* Separated out issue of normalising TCP's bit rate from issue of
preference to control packets (Section 2.3).
* Divided up Congestion Measurement section for clarity,
including new material on fixed size packet buffers and buffer
carving (Section 4.1.1 & Section 6.2.1) and on congestion
measurement in wireless link technologies without queues
(Section 4.2).
* Added section on 'Making Transports Robust against Control Appendix C. Byte-mode Drop Complicates Policing Congestion
Packet Losses' (Section 6.2.3) with existing & new material Response . . . . . . . . . . . . . . . . . . . . . . 35
included. Appendix D. Changes from Previous Versions . . . . . . . . . . . 36
* Added tabulated results of vendor survey on byte-mode drop 1. Introduction
variant of RED (Table 2).
* When notifying congestion, the problem of how (and whether) to take
packet sizes into account has exercised the minds of researchers and
practitioners for as long as active queue management (AQM) has been
discussed. Indeed, one reason AQM was originally introduced was to
reduce the lock-out effects that small packets can have on large
packets in drop-tail queues. This memo aims to state the principles
we should be using and to come to conclusions on what these
principles will mean for future protocol design, taking into account
the deployments we have already.
From -00 to -01: The byte vs. packet dilemma arises at three stages in the congestion
notification process:
* Clarified applicability to drop as well as ECN. Measuring congestion: When the congested resource decides locally to
measure how congested it is. (Should the queue measure its length
in bytes or packets?);
* Highlighted DoS vulnerability. Coding congestion notification into the wire protocol: When the
congested resource decides whether to notify the level of
congestion on each particular packet. (When a queue considers
whether to notify congestion by dropping or marking a particular
packet, should its decision depend on the byte-size of the
particular packet being dropped or marked?);
* Emphasised that drop-tail suffers from similar problems to Decoding congestion notification from the wire protocol: When the
byte-mode drop, so only byte-mode drop should be turned off, transport interprets the notification in order to decide how much
not RED itself. to respond to congestion. (Should the transport take into account
the byte-size of each missing or marked packet?).
* Clarified the original apparent motivations for recommending Consensus has emerged over the years concerning the first stage:
byte-mode drop included protecting SYNs and pure ACKs more than whether queues are measured in bytes or packets, termed byte-mode
equalising the bit rates of TCPs with different segment sizes. queue measurement or packet-mode queue measurement. This memo
Removed some conjectured motivations. records this consensus in the RFC Series. In summary the choice
solely depends on whether the resource is congested by bytes or
packets.
* Added support for updates to TCP in progress (ackcc & ecn-syn- The controversy is mainly around the last two stages to do with
ack). encoding congestion notification into packets: whether to allow for
the size of the specific packet notifying congestion i) when the
network encodes or ii) when the transport decodes the congestion
notification.
* Updated survey results with newly arrived data. Currently, the RFC series is silent on this matter other than a paper
trail of advice referenced from [RFC2309], which conditionally
recommends byte-mode (packet-size dependent) drop [pktByteEmail].
The primary purpose of this memo is to build a definitive consensus
against such deliberate preferential treatment for small packets in
AQM algorithms and to record this advice within the RFC series.
Fortunately all the implementers who responded to our survey
(Section 3.2.4) have not followed the earlier advice, so the
consensus this memo argues for seems to already exist in
implementations.
* Pulled all recommendations together into the conclusions. The primary conclusion of this memo is that packet size should be
taken into account when transports read congestion indications, not
when network equipment writes them. Reducing drop of small packets
has some tempting advantages: i) it drops less control packets, which
tend to be small and ii) it makes TCP's bit-rate less dependent on
packet size. However, there are ways of addressing these issues at
the transport layer, rather than reverse engineering network
forwarding to fix specific transport problems.
* Moved some detailed points into two additional appendices and a The second conclusion is that network layer algorithms like the byte-
note. mode packet drop variant of RED should not be used to drop fewer
small packets, because that creates a perverse incentive for
transports to use tiny segments, consequently also opening up a DoS
vulnerability.
* Considerable clarifications throughout. This memo is initially concerned with how we should correctly scale
congestion control functions with packet size for the long term. But
it also recognises that expediency may be necessary to deal with
existing widely deployed protocols that don't live up to the long
term goal. It turns out that the 'correct' variant of RED to deploy
seems to be the one everyone has deployed, and no-one who responded
to our survey has implemented the other variant. However, at the
transport layer, TCP congestion control is a widely deployed protocol
that we argue doesn't scale correctly with packet size. To date this
hasn't been a significant problem because most TCPs have been used
with similar packet sizes. But, as we design new congestion
controls, we should build in scaling with packet size rather than
assuming we should follow TCP's example.
* Updated references This memo continues as follows. Terminology and scoping are
discussed next, and the reasons to make the recommendations presented
in this memo now are given in Section 1.2. Motivating arguments for
our advice are given in Section 2. We then survey the advice given
previously in the RFC series, the research literature and the
deployed legacy (Section 3) before listing outstanding issues
(Section 4) that will need resolution both to inform future protocols
designs and to handle legacy. We then give concrete recommendations
for the way forward in (Section 5). We finally give security
considerations in Section 6. The interested reader can also find
further discussions about the theme of byte vs. packet in the
appendices.
1. Introduction This memo intentionally includes a non-negligible amount of material
on the subject. A busy reader can jump right into Section 5 to read
a summary of the recommendations for the Internet community.
When notifying congestion, the problem of how (and whether) to take 1.1. Terminology and Scoping
packet sizes into account has exercised the minds of researchers and
practitioners for as long as active queue management (AQM) has been
discussed. Indeed, one reason AQM was originally introduced was to
reduce the lock-out effects that small packets can have on large
packets in drop-tail queues. This memo aims to state the principles
we should be using and to come to conclusions on what these
principles will mean for future protocol design, taking into account
the deployments we have already.
Note that the byte vs. packet dilemma concerns congestion The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
notification irrespective of whether it is signalled implicitly by "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this
drop or using explicit congestion notification (ECN [RFC3168] or PCN document are to be interpreted as described in [RFC2119].
[I-D.ietf-pcn-marking-behaviour]). Throughout this document, unless
clear from the context, the term marking will be used to mean
notifying congestion explicitly, while congestion notification will
be used to mean notifying congestion either implicitly by drop or
explicitly by marking.
If the load on a resource depends on the rate at which packets Congestion Notification: Rather than aim to achieve what many have
arrive, it is called packet-congestible. If the load depends on the tried and failed, this memo will not try to define congestion. It
rate at which bits arrive it is called bit-congestible. will give a working definition of what congestion notification
should be taken to mean for this document. Congestion
notification is a changing signal that aims to communicate the
ratio E/L. E is the instantaneous excess load offered to a
resource that it is either incapable of serving or unwilling to
serve. L is the instantaneous offered load.
Examples of packet-congestible resources are route look-up engines The phrase `unwilling to serve' is added, because AQM systems
and firewalls, because load depends on how many packet headers they (e.g. RED, PCN [RFC5670]) set a virtual limit smaller than the
have to process. Examples of bit-congestible resources are actual limit to the resource, then notify when this virtual limit
transmission links, radio power and most buffer memory, because the is exceeded in order to avoid congestion of the actual capacity.
load depends on how many bits they have to transmit or store. Some
machine architectures use fixed size packet buffers, so buffer memory
in these cases is packet-congestible (see Section 4.1.1).
Note that information is generally processed or transmitted with a Note that the denominator is offered load, not capacity.
minimum granularity greater than a bit (e.g. octets). The Therefore congestion notification is a real number bounded by the
appropriate granularity for the resource in question SHOULD be used, range [0,1]. This ties in with the most well-understood measure
but for the sake of brevity we will talk in terms of bytes in this of congestion notification: drop fraction (often loosely called
memo. loss rate). It also means that congestion has a natural
interpretation as a probability; the probability of offered
traffic not being served (or being marked as at risk of not being
served). Appendix A describes a further incidental benefit that
arises from using load as the denominator of congestion
notification.
Resources may be congestible at higher levels of granularity than Explicit and Implicit Notification: The byte vs. packet dilemma
packets, for instance stateful firewalls are flow-congestible and concerns congestion notification irrespective of whether it is
call-servers are session-congestible. This memo focuses on signalled implicitly by drop or using explicit congestion
congestion of connectionless resources, but the same principles may notification (ECN [RFC3168] or PCN [RFC5670]). Throughout this
be applicable for congestion notification protocols controlling per- document, unless clear from the context, the term marking will be
flow and per-session processing or state. used to mean notifying congestion explicitly, while congestion
notification will be used to mean notifying congestion either
implicitly by drop or explicitly by marking.
The byte vs. packet dilemma arises at three stages in the congestion Bit-congestible vs. Packet-congestible: If the load on a resource
notification process: depends on the rate at which packets arrive, it is called packet-
congestible. If the load depends on the rate at which bits arrive
it is called bit-congestible.
Measuring congestion When the congested resource decides locally how Examples of packet-congestible resources are route look-up engines
to measure how congested it is. (Should the queue be measured in and firewalls, because load depends on how many packet headers
bytes or packets?); they have to process. Examples of bit-congestible resources are
transmission links, radio power and most buffer memory, because
the load depends on how many bits they have to transmit or store.
Some machine architectures use fixed size packet buffers, so
buffer memory in these cases is packet-congestible (see
Section 3.1.1).
Coding congestion notification into the wire protocol: When the Currently a design goal of network processing equipment such as
congested resource decides how to notify the level of congestion. routers and firewalls is to keep packet processing uncongested
(Should the level of notification depend on the byte-size of each even under worst case bit rates with minimum packet sizes.
particular packet carrying the notification?); Therefore, packet-congestion is currently rare, but there is no
guarantee that it will not become common with future technology
trends.
Decoding congestion notification from the wire protocol: When the Note that information is generally processed or transmitted with a
transport interprets the notification. (Should the byte-size of a minimum granularity greater than a bit (e.g. octets). The
missing or marked packet be taken into account?). appropriate granularity for the resource in question should be
used, but for the sake of brevity we will talk in terms of bytes
in this memo.
In RED, whether to use packets or bytes when measuring queues is Coarser granularity: Resources may be congestible at higher levels
called packet-mode or byte-mode queue measurement. This choice is of granularity than packets, for instance stateful firewalls are
now fairly well understood but is included in Section 4 to document flow-congestible and call-servers are session-congestible. This
it in the RFC series. memo focuses on congestion of connectionless resources, but the
same principles may be applicable for congestion notification
protocols controlling per-flow and per-session processing or
state.
The controversy is mainly around the other two stages: whether to RED Terminology: In RED, whether to use packets or bytes when
allow for packet size when the network codes or when the transport measuring queues is respectively called packet-mode or byte-mode
decodes congestion notification. In RED, the variant that reduces queue measurement. And if the probability of dropping a packet
drop probability for packets based on their size in bytes is called depends on its byte-size it is called byte-mode drop, whereas if
byte-mode drop, while the variant that doesn't is called packet mode the drop probability is independent of a packet's byte-size it is
drop. Whether queues are measured in bytes or packets is an called packet-mode drop.
orthogonal choice, termed byte-mode queue measurement or packet-mode
queue measurement.
Currently, the RFC series is silent on this matter other than a paper 1.2. Why now?
trail of advice referenced from [RFC2309], which conditionally
recommends byte-mode (packet-size dependent) drop [pktByteEmail].
However, all the implementers who responded to our survey
(Section 6.2.4) have not followed this advice. The primary purpose
of this memo is to build a definitive consensus against deliberate
preferential treatment for small packets in AQM algorithms and to
record this advice within the RFC series.
Now is a good time to discuss whether fairness between different Now is a good time to discuss whether fairness between different
sized packets would best be implemented in the network layer, or at sized packets would best be implemented in the network layer, or at
the transport, for a number of reasons: the transport, for a number of reasons:
1. The packet vs. byte issue requires speedy resolution because the 1. The packet vs. byte issue requires speedy resolution because the
IETF pre-congestion notification (PCN) working group is about to IETF pre-congestion notification (PCN) working group is
standardise the external behaviour of a PCN congestion standardising the external behaviour of a PCN congestion
notification (AQM) algorithm [I-D.ietf-pcn-marking-behaviour]; notification (AQM) algorithm [RFC5670];
2. [RFC2309] says RED may either take account of packet size or not 2. [RFC2309] says RED may either take account of packet size or not
when dropping, but gives no recommendation between the two, when dropping, but gives no recommendation between the two,
referring instead to advice on the performance implications in an referring instead to advice on the performance implications in an
email [pktByteEmail], which recommends byte-mode drop. Further, email [pktByteEmail], which recommends byte-mode drop. Further,
just before RFC2309 was issued, an addendum was added to the just before RFC2309 was issued, an addendum was added to the
archived email that revisited the issue of packet vs. byte-mode archived email that revisited the issue of packet vs. byte-mode
drop in its last para, making the recommendation less clear-cut; drop in its last paragraph, making the recommendation less clear-
cut;
3. Without the present memo, the only advice in the RFC series on 3. Without the present memo, the only advice in the RFC series on
packet size bias in AQM algorithms would be a reference to an packet size bias in AQM algorithms would be a reference to an
archived email in [RFC2309] (including an addendum at the end of archived email in [RFC2309] (including an addendum at the end of
the email to correct the original). the email to correct the original).
4. The IRTF Internet Congestion Control Research Group (ICCRG) 4. The IRTF Internet Congestion Control Research Group (ICCRG)
recently took on the challenge of building consensus on what recently took on the challenge of building consensus on what
common congestion control support should be required from network common congestion control support should be required from network
forwarding functions in future forwarding functions in future [I-D.irtf-iccrg-welzl]. The wider
[I-D.irtf-iccrg-welzl-congestion-control-open-research]. The Internet community needs to discuss whether the complexity of
wider Internet community needs to discuss whether the complexity adjusting for packet size should be in the network or in
of adjusting for packet size should be in the network or in
transports; transports;
5. Given there are many good reasons why larger path max 5. Given there are many good reasons why larger path max
transmission units (PMTUs) would help solve a number of scaling transmission units (PMTUs) would help solve a number of scaling
issues, we don't want to create any bias against large packets issues, we don't want to create any bias against large packets
that is greater than their true cost; that is greater than their true cost;
6. The IETF has started to consider the question of fairness between 6. The IETF has started to consider the question of fairness between
flows that use different packet sizes (e.g. in the small-packet flows that use different packet sizes (e.g. in the small-packet
variant of TCP-friendly rate control, TFRC-SP [RFC4828]). Given variant of TCP-friendly rate control, TFRC-SP [RFC4828]). Given
transports with different packet sizes, if we don't decide transports with different packet sizes, if we don't decide
whether the network or the transport should allow for packet whether the network or the transport should allow for packet
size, it will be hard if not impossible to design any transport size, it will be hard if not impossible to design any transport
protocol so that its bit-rate relative to other transports meets protocol so that its bit-rate relative to other transports meets
design guidelines [RFC5033] (Note however that, if the concern design guidelines [RFC5033] (Note however that, if the concern
were fairness between users, rather than between flows were fairness between users, rather than between flows
[Rate_fair_Dis], relative rates between flows would have to come [Rate_fair_Dis], relative rates between flows would have to come
under run-time control rather than being embedded in protocol under run-time control rather than being embedded in protocol
designs). designs).
This memo is initially concerned with how we should correctly scale
congestion control functions with packet size for the long term. But
it also recognises that expediency may be necessary to deal with
existing widely deployed protocols that don't live up to the long
term goal. It turns out that the 'correct' variant of RED to deploy
seems to be the one everyone has deployed, and no-one who responded
to our survey has implemented the other variant. However, at the
transport layer, TCP congestion control is a widely deployed protocol
that we argue doesn't scale correctly with packet size. To date this
hasn't been a significant problem because most TCPs have been used
with similar packet sizes. But, as we design new congestion
controls, we should build in scaling with packet size rather than
assuming we should follow TCP's example.
Motivating arguments for our advice are given next in Section 2.
Then the body of the memo starts from first principles, defining
congestion notification in Section 3 then determining the correct way
to measure congestion (Section 4) and to design an idealised
congestion notification protocol (Section 5). It then surveys the
advice given previously in the RFC series, the research literature
and the deployed legacy (Section 6) before listing outstanding issues
(Section 7) that will need resolution both to achieve the ideal
protocol and to handle legacy. After discussing security
considerations (Section 8) strong recommendations for the way forward
are given in the conclusions (Section 9).
1.1. Requirements Notation
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].
2. Motivating Arguments 2. Motivating Arguments
2.1. Scaling Congestion Control with Packet Size 2.1. Scaling Congestion Control with Packet Size
There are two ways of interpreting a dropped or marked packet. It There are two ways of interpreting a dropped or marked packet. It
can either be considered as a single loss event or as loss/marking of can either be considered as a single loss event or as loss/marking of
the bytes in the packet. Here we try to design a test to see which the bytes in the packet. Here we try to design a test to see which
approach scales with packet size. approach scales with packet size.
Given bit-congestible is the more common case, consider a bit- Given bit-congestible is the more common case (see Section 1.1),
congestible link shared by many flows, so that each busy period tends consider a bit-congestible link shared by many flows, so that each
to cause packets to be lost from different flows. The test compares busy period tends to cause packets to be lost from different flows.
two identical scenarios with the same applications, the same numbers The test compares two identical scenarios with the same applications,
of sources and the same load. But the sources break the load into the same numbers of sources and the same load. But the sources break
large packets in one scenario and small packets in the other. Of the load into large packets in one scenario and small packets in the
course, because the load is the same, there will be proportionately other. Of course, because the load is the same, there will be
more packets in the small packet case. proportionately more packets in the small packet case.
The test of whether a congestion control scales with packet size is The test of whether a congestion control scales with packet size is
that it should respond in the same way to the same congestion that it should respond in the same way to the same congestion
excursion, irrespective of the size of the packets that the bytes excursion, irrespective of the size of the packets that the bytes
causing congestion happen to be broken down into. causing congestion happen to be broken down into.
A bit-congestible queue suffering a congestion excursion has to drop A bit-congestible queue suffering a congestion excursion has to drop
or mark the same excess bytes whether they are in a few large packets or mark the same excess bytes whether they are in a few large packets
or many small packets. So for the same congestion excursion, the or many small packets. So for the same congestion excursion, the
same amount of bytes have to be shed to get the load back to its same amount of bytes have to be shed to get the load back to its
skipping to change at page 10, line 38 skipping to change at page 9, line 50
"depends on the dominant end-to-end congestion control mechanisms". "depends on the dominant end-to-end congestion control mechanisms".
But we argue the network layer should not be optimised for whatever But we argue the network layer should not be optimised for whatever
transport is predominant. transport is predominant.
TCP congestion control ensures that flows competing for the same TCP congestion control ensures that flows competing for the same
resource each maintain the same number of segments in flight, resource each maintain the same number of segments in flight,
irrespective of segment size. So under similar conditions, flows irrespective of segment size. So under similar conditions, flows
with different segment sizes will get different bit rates. But even with different segment sizes will get different bit rates. But even
though reducing the drop probability of small packets helps ensure though reducing the drop probability of small packets helps ensure
TCPs with different packet sizes will achieve similar bit rates, we TCPs with different packet sizes will achieve similar bit rates, we
argue this should be achieved in TCP itself, not in the network. argue this correction should be made to TCP itself, not to the
network in order to fix one transport, no matter how prominent it is.
Effectively, favouring small packets is reverse engineering of the Effectively, favouring small packets is reverse engineering of the
network layer around TCP, contrary to the excellent advice in network layer around TCP, contrary to the excellent advice in
[RFC3426], which asks designers to question "Why are you proposing a [RFC3426], which asks designers to question "Why are you proposing a
solution at this layer of the protocol stack, rather than at another solution at this layer of the protocol stack, rather than at another
layer?" layer?"
2.2. Avoiding Perverse Incentives to (ab)use Smaller Packets 2.2. Avoiding Perverse Incentives to (ab)use Smaller Packets
Increasingly, it is being recognised that a protocol design must take Increasingly, it is being recognised that a protocol design must take
care not to cause unintended consequences by giving the parties in care not to cause unintended consequences by giving the parties in
the protocol exchange perverse incentives [Evol_cc][RFC3426]. Again, the protocol exchange perverse incentives [Evol_cc][RFC3426]. Again,
imagine a scenario where the same bit rate of packets will contribute imagine a scenario where the same bit rate of packets will contribute
the same to congestion of a link irrespective of whether it is sent the same to bit-congestion of a link irrespective of whether it is
as fewer larger packets or more smaller packets. A protocol design sent as fewer larger packets or more smaller packets. A protocol
that caused larger packets to be more likely to be dropped than design that caused larger packets to be more likely to be dropped
smaller ones would be dangerous in this case: than smaller ones would be dangerous in this case:
Normal transports: Even if a transport is not actually malicious, if
it finds small packets go faster, over time it will tend to act in
its own interest and use them. Queues that give advantage to
small packets create an evolutionary pressure for transports to
send at the same bit-rate but break their data stream down into
tiny segments to reduce their drop rate. Encouraging a high
volume of tiny packets might in turn unnecessarily overload a
completely unrelated part of the system, perhaps more limited by
header-processing than bandwidth.
Malicious transports: A queue that gives an advantage to small Malicious transports: A queue that gives an advantage to small
packets can be used to amplify the force of a flooding attack. By packets can be used to amplify the force of a flooding attack. By
sending a flood of small packets, the attacker can get the queue sending a flood of small packets, the attacker can get the queue
to discard more traffic in large packets, allowing more attack to discard more traffic in large packets, allowing more attack
traffic to get through to cause further damage. Such a queue traffic to get through to cause further damage. Such a queue
allows attack traffic to have a disproportionately large effect on allows attack traffic to have a disproportionately large effect on
regular traffic without the attacker having to do much work. regular traffic without the attacker having to do much work.
Note that, although the byte-mode drop variant of RED amplifies Note that, although the byte-mode drop variant of RED amplifies
small packet attacks, drop-tail queues amplify small packet small packet attacks, drop-tail queues amplify small packet
attacks even more (see Security Considerations in Section 8). attacks even more (see Security Considerations in Section 6).
Wherever possible neither should be used. Wherever possible neither should be used.
Normal transports: Even if a transport is not malicious, if it finds
small packets go faster, it will tend to act in its own interest
and use them. Queues that give advantage to small packets create
an evolutionary pressure for transports to send at the same bit-
rate but break their data stream down into tiny segments to reduce
their drop rate. Encouraging a high volume of tiny packets might
in turn unnecessarily overload a completely unrelated part of the
system, perhaps more limited by header-processing than bandwidth.
Imagine two unresponsive flows arrive at a bit-congestible Imagine two unresponsive flows arrive at a bit-congestible
transmission link each with the same bit rate, say 1Mbps, but one transmission link each with the same bit rate, say 1Mbps, but one
consists of 1500B and the other 60B packets, which are 25x smaller. consists of 1500B and the other 60B packets, which are 25x smaller.
Consider a scenario where gentle RED [gentle_RED] is used, along with Consider a scenario where gentle RED [gentle_RED] is used, along with
the variant of RED we advise against, i.e. where the RED algorithm is the variant of RED we advise against, i.e. where the RED algorithm is
configured to adjust the drop probability of packets in proportion to configured to adjust the drop probability of packets in proportion to
each packet's size (byte mode packet drop). In this case, if RED each packet's size (byte mode packet drop). In this case, if RED
drops 25% of the larger packets, it will aim to drop 1% of the drops 25% of the larger packets, it will aim to drop 1% of the
smaller packets (but in practice it may drop more as congestion smaller packets (but in practice it may drop more as congestion
increases [RFC4828](S.B.4)[Note_Variation]). Even though both flows increases [RFC4828](S.B.4)). Even though both flows arrive with the
arrive with the same bit rate, the bit rate the RED queue aims to same bit rate, the bit rate the RED queue aims to pass to the line
pass to the line will be 750k for the flow of larger packet but 990k will be 750k for the flow of larger packet but 990k for the smaller
for the smaller packets (but because of rate variation it will be packets (but because of rate variation it will be less than this
less than this target). target).
It can be seen that this behaviour reopens the same denial of service It can be seen that this behaviour reopens the same denial of service
vulnerability that drop tail queues offer to floods of small packet, vulnerability that drop tail queues offer to floods of small packet,
though not necessarily as strongly (see Section 8). though not necessarily as strongly (see Section 6).
2.3. Small != Control 2.3. Small != Control
It is tempting to drop small packets with lower probability to It is tempting to drop small packets with lower probability to
improve performance, because many control packets are small (TCP SYNs improve performance, because many control packets are small (TCP SYNs
& ACKs, DNS queries & responses, SIP messages, HTTP GETs, etc) and & ACKs, DNS queries & responses, SIP messages, HTTP GETs, etc) and
dropping fewer control packets considerably improves performance. dropping fewer control packets considerably improves performance.
However, we must not give control packets preference purely by virtue However, we must not give control packets preference purely by virtue
of their smallness, otherwise it is too easy for any data source to of their smallness, otherwise it is too easy for any data source to
get the same preferential treatment simply by sending data in smaller get the same preferential treatment simply by sending data in smaller
packets. Again we should not create perverse incentives to favour packets. Again we should not create perverse incentives to favour
small packets rather than to favour control packets, which is what we small packets rather than to favour control packets, which is what we
intend. intend.
Just because many control packets are small does not mean all small Just because many control packets are small does not mean all small
packets are control packets. packets are control packets.
So again, rather than fix these problems in the network layer, we So again, rather than fix these problems in the network layer, we
argue that the transport should be made more robust against losses of argue that the transport should be made more robust against losses of
control packets (see 'Making Transports Robust against Control Packet control packets (see 'Making Transports Robust against Control Packet
Losses' in Section 6.2.3). Losses' in Section 3.2.3).
2.4. Implementation Efficiency 2.4. Implementation Efficiency
Allowing for packet size at the transport rather than in the network Allowing for packet size at the transport rather than in the network
ensures that neither the network nor the transport needs to do a ensures that neither the network nor the transport needs to do a
multiply operation--multiplication by packet size is effectively multiply operation--multiplication by packet size is effectively
achieved as a repeated add when the transport adds to its count of achieved as a repeated add when the transport adds to its count of
marked bytes as each congestion event is fed to it. This isn't a marked bytes as each congestion event is fed to it. This isn't a
principled reason in itself, but it is a happy consequence of the principled reason in itself, but it is a happy consequence of the
other principled reasons. other principled reasons.
3. Working Definition of Congestion Notification 3. The State of the Art
Rather than aim to achieve what many have tried and failed, this memo The original 1993 paper on RED [RED93] proposed two options for the
will not try to define congestion. It will give a working definition RED active queue management algorithm: packet mode and byte mode.
of what congestion notification should be taken to mean for this Packet mode measured the queue length in packets and dropped (or
document. Congestion notification is a changing signal that aims to marked) individual packets with a probability independent of their
communicate the ratio E/L, where E is the instantaneous excess load size. Byte mode measured the queue length in bytes and marked an
offered to a resource that it cannot (or would not) serve and L is individual packet with probability in proportion to its size
the instantaneous offered load. (relative to the maximum packet size). In the paper's outline of
further work, it was stated that no recommendation had been made on
whether the queue size should be measured in bytes or packets, but
noted that the difference could be significant.
The phrase `would not serve' is added, because AQM systems (e.g. When RED was recommended for general deployment in 1998 [RFC2309],
RED, PCN [I-D.ietf-pcn-marking-behaviour]) use a virtual capacity the two modes were mentioned implying the choice between them was a
smaller than actual capacity, then notify congestion of this virtual question of performance, referring to a 1997 email [pktByteEmail] for
capacity in order to avoid congestion of the actual capacity. advice on tuning. This email clarified that there were in fact two
orthogonal choices: whether to measure queue length in bytes or
packets (Section 3.1 below) and whether the drop probability of an
individual packet should depend on its own size (Section 3.2 below).
Note that the denominator is offered load, not capacity. Therefore 3.1. Congestion Measurement: Status
congestion notification is a real number bounded by the range [0,1].
This ties in with the most well-understood measure of congestion
notification: drop fraction (often loosely called loss rate). It
also means that congestion has a natural interpretation as a
probability; the probability of offered traffic not being served (or
being marked as at risk of not being served). Appendix B describes a
further incidental benefit that arises from using load as the
denominator of congestion notification.
4. Congestion Measurement The choice of which metric to use to measure queue length was left
open in RFC2309. It is now well understood that queues for bit-
congestible resources should be measured in bytes, and queues for
packet-congestible resources should be measured in packets.
4.1. Congestion Measurement by Queue Length Where buffers are not configured or legacy buffers cannot be
configured to the above guideline, we do not have to make allowances
for such legacy in future protocol design. If a bit-congestible
buffer is measured in packets, the operator will have set the
thresholds mindful of a typical mix of packets sizes. Any AQM
algorithm on such a buffer will be oversensitive to high proportions
of small packets, e.g. a DoS attack, and undersensitive to high
proportions of large packets. But an operator can safely keep such a
legacy buffer because any undersensitivity during unusual traffic
mixes cannot lead to congestion collapse given the buffer will
eventually revert to tail drop, discarding proportionately more large
packets.
Queue length is usually the most correct and simplest way to measure Some modern queue implementations give a choice for setting RED's
congestion of a resource. To avoid the pathological effects of drop thresholds in byte-mode or packet-mode. This may merely be an
tail, an AQM function can then be used to transform queue length into administrator-interface preference, not altering how the queue itself
the probability of dropping or marking a packet (e.g. RED's is measured but on some hardware it does actually change the way it
piecewise linear function between thresholds). If the resource is measures its queue. Whether a resource is bit-congestible or packet-
bit-congestible, the length of the queue SHOULD be measured in bytes. congestible is a property of the resource, so an admin should not
If the resource is packet-congestible, the length of the queue SHOULD ever need to, or be able to, configure the way a queue measures
be measured in packets. No other choice makes sense, because the itself.
number of packets waiting in the queue isn't relevant if the resource
gets congested by bytes and vice versa. We discuss the implications
on RED's byte mode and packet mode for measuring queue length in
Section 6.
4.1.1. Fixed Size Packet Buffers We believe the question of whether to measure queues in bytes or
packets is fairly well understood these days. The only outstanding
issues concern how to measure congestion when the queue is bit
congestible but the resource is packet congestible or vice versa.
There is no controversy over what should be done. It's just you have
to be an expert in probability to work out what should be done
(summarised in the following section) and, even if you have, it's not
always easy to find a practical algorithm to implement it.
3.1.1. Fixed Size Packet Buffers
Some, mostly older, queuing hardware sets aside fixed sized buffers Some, mostly older, queuing hardware sets aside fixed sized buffers
in which to store each packet in the queue. Also, with some in which to store each packet in the queue. Also, with some
hardware, any fixed sized buffers not completely filled by a packet hardware, any fixed sized buffers not completely filled by a packet
are padded when transmitted to the wire. If we imagine a theoretical are padded when transmitted to the wire. If we imagine a theoretical
forwarding system with both queuing and transmission in fixed, MTU- forwarding system with both queuing and transmission in fixed, MTU-
sized units, it should clearly be treated as packet-congestible, sized units, it should clearly be treated as packet-congestible,
because the queue length in packets would be a good model of because the queue length in packets would be a good model of
congestion of the lower layer link. congestion of the lower layer link.
skipping to change at page 14, line 5 skipping to change at page 13, line 32
largely dependent on the byte-size of packets but buffers of one MTU largely dependent on the byte-size of packets but buffers of one MTU
per packet, it should strictly require a more complex algorithm to per packet, it should strictly require a more complex algorithm to
determine the probability of congestion. It should be treated as two determine the probability of congestion. It should be treated as two
resources in sequence, where the sum of the byte-sizes of the packets resources in sequence, where the sum of the byte-sizes of the packets
within each packet buffer models congestion of the line while the within each packet buffer models congestion of the line while the
length of the queue in packets models congestion of the queue. Then length of the queue in packets models congestion of the queue. Then
the probability of congesting the forwarding buffer would be a the probability of congesting the forwarding buffer would be a
conditional probability--conditional on the previously calculated conditional probability--conditional on the previously calculated
probability of congesting the line. probability of congesting the line.
However, in systems that use fixed size buffers, it is unusual for In systems that use fixed size buffers, it is unusual for all the
all the buffers used by an interface to be the same size. Typically buffers used by an interface to be the same size. Typically pools of
pools of different sized buffers are provided (Cisco uses the term different sized buffers are provided (Cisco uses the term 'buffer
'buffer carving' for the process of dividing up memory into these carving' for the process of dividing up memory into these pools
pools [IOSArch]). Usually, if the pool of small buffers is [IOSArch]). Usually, if the pool of small buffers is exhausted,
exhausted, arriving small packets can borrow space in the pool of arriving small packets can borrow space in the pool of large buffers,
large buffers, but not vice versa. However, it is easier to work out but not vice versa. However, it is easier to work out what should be
what should be done if we temporarily set aside the possibility of done if we temporarily set aside the possibility of such borrowing.
such borrowing. Then, with fixed pools of buffers for different Then, with fixed pools of buffers for different sized packets and no
sized packets and no borrowing, the size of each pool and the current borrowing, the size of each pool and the current queue length in each
queue length in each pool would both be measured in packets. So an pool would both be measured in packets. So an AQM algorithm would
AQM algorithm would have to maintain the queue length for each pool, have to maintain the queue length for each pool, and judge whether to
and judge whether to drop/mark a packet of a particular size by drop/mark a packet of a particular size by looking at the pool for
looking at the pool for packets of that size and using the length (in packets of that size and using the length (in packets) of its queue.
packets) of its queue.
We now return to the issue we temporarily set aside: small packets We now return to the issue we temporarily set aside: small packets
borrowing space in larger buffers. In this case, the only difference borrowing space in larger buffers. In this case, the only difference
is that the pools for smaller packets have a maximum queue size that is that the pools for smaller packets have a maximum queue size that
includes all the pools for larger packets. And every time a packet includes all the pools for larger packets. And every time a packet
takes a larger buffer, the current queue size has to be incremented takes a larger buffer, the current queue size has to be incremented
for all queues in the pools of buffers less than or equal to the for all queues in the pools of buffers less than or equal to the
buffer size used. buffer size used.
We will return to borrowing of fixed sized buffers when we discuss We will return to borrowing of fixed sized buffers when we discuss
biasing the drop/marking probability of a specific packet because of biasing the drop/marking probability of a specific packet because of
its size in Section 6.2.1. But here we can give a simple summary of its size in Section 3.2.1. But here we can give a simple summary of
the present discussion on how to measure the length of queues of the present discussion on how to measure the length of queues of
fixed buffers: no matter how complicated the scheme is, ultimately fixed buffers: no matter how complicated the scheme is, ultimately
any fixed buffer system will need to measure its queue length in any fixed buffer system will need to measure its queue length in
packets not bytes. packets not bytes.
4.2. Congestion Measurement without a Queue 3.1.2. Congestion Measurement without a Queue
AQM algorithms are nearly always described assuming there is a queue AQM algorithms are nearly always described assuming there is a queue
for a congested resource and the algorithm can use the queue length for a congested resource and the algorithm can use the queue length
to determine the probability that it will drop or mark each packet. to determine the probability that it will drop or mark each packet.
But not all congested resources lead to queues. For instance, But not all congested resources lead to queues. For instance,
wireless spectrum is bit-congestible (for a given coding scheme), wireless spectrum is bit-congestible (for a given coding scheme),
because interference increases with the rate at which bits are because interference increases with the rate at which bits are
transmitted. But wireless link protocols do not always maintain a transmitted. But wireless link protocols do not always maintain a
queue that depends on spectrum interference. Similarly, power queue that depends on spectrum interference. Similarly, power
limited resources are also usually bit-congestible if energy is limited resources are also usually bit-congestible if energy is
primarily required for transmission rather than header processing, primarily required for transmission rather than header processing,
but it is rare for a link protocol to build a queue as it approaches but it is rare for a link protocol to build a queue as it approaches
maximum power. maximum power.
However, AQM algorithms don't require a queue in order to work. For Nonetheless, AQM algorithms do not require a queue in order to work.
instance spectrum congestion can be modelled by signal quality using For instance spectrum congestion can be modelled by signal quality
target bit-energy-to-noise-density ratio. And, to model radio power using target bit-energy-to-noise-density ratio. And, to model radio
exhaustion, transmission power levels can be measured and compared to power exhaustion, transmission power levels can be measured and
the maximum power available. [ECNFixedWireless] proposes a practical compared to the maximum power available. [ECNFixedWireless] proposes
and theoretically sound way to combine congestion notification for a practical and theoretically sound way to combine congestion
different bit-congestible resources at different layers along an end notification for different bit-congestible resources at different
to end path, whether wireless or wired, and whether with or without layers along an end to end path, whether wireless or wired, and
queues. whether with or without queues.
5. Idealised Wire Protocol Coding
We will start by inventing an idealised congestion notification
protocol before discussing how to make it practical. The idealised
protocol is shown to be correct using examples in Appendix A.
Congestion notification involves the congested resource coding a
congestion notification signal into the packet stream and the
transports decoding it. The idealised protocol uses two different
(imaginary) fields in each datagram to signal congestion: one for
byte congestion and one for packet congestion.
We are not saying two ECN fields will be needed (and we are not
saying that somehow a resource should be able to drop a packet in one
of two different ways so that the transport can distinguish which
sort of drop it was!). These two congestion notification channels
are just a conceptual device. They allow us to defer having to
decide whether to distinguish between byte and packet congestion when
the network resource codes the signal or when the transport decodes
it.
However, although this idealised mechanism isn't intended for
implementation, we do want to emphasise that we may need to find a
way to implement it, because it could become necessary to somehow
distinguish between bit and packet congestion [RFC3714]. Currently a
design goal of network processing equipment such as routers and
firewalls is to keep packet processing uncongested even under worst
case bit rates with minimum packet sizes. Therefore, packet-
congestion is currently rare, but there is no guarantee that it will
not become common with future technology trends.
The idealised wire protocol is given below. It accounts for packet
sizes at the transport layer, not in the network, and then only in
the case of bit-congestible resources. This avoids the perverse
incentive to send smaller packets and the DoS vulnerability that
would otherwise result if the network were to bias towards them (see
the motivating argument about avoiding perverse incentives in
Section 2.2):
1. A packet-congestible resource trying to code congestion level p_p
into a packet stream should mark the idealised `packet
congestion' field in each packet with probability p_p
irrespective of the packet's size. The transport should then
take a packet with the packet congestion field marked to mean
just one mark, irrespective of the packet size.
2. A bit-congestible resource trying to code time-varying byte-
congestion level p_b into a packet stream should mark the `byte
congestion' field in each packet with probability p_b, again
irrespective of the packet's size. Unlike before, the transport
should take a packet with the byte congestion field marked to
count as a mark on each byte in the packet.
The worked examples in Appendix A show that transports can extract
sufficient and correct congestion notification from these protocols
for cases when two flows with different packet sizes have matching
bit rates or matching packet rates. Examples are also given that mix
these two flows into one to show that a flow with mixed packet sizes
would still be able to extract sufficient and correct information.
Sufficient and correct congestion information means that there is
sufficient information for the two different types of transport
requirements:
Ratio-based: Established transport congestion controls like TCP's
[RFC5681] aim to achieve equal segment rates per RTT through the
same bottleneck--TCP friendliness [RFC3448]. They work with the
ratio of dropped to delivered segments (or marked to unmarked
segments in the case of ECN). The example scenarios show that
these ratio-based transports are effectively the same whether
counting in bytes or packets, because the units cancel out.
(Incidentally, this is why TCP's bit rate is still proportional to
packet size even when byte-counting is used, as recommended for
TCP in [RFC5681], mainly for orthogonal security reasons.)
Absolute-target-based: Other congestion controls proposed in the
research community aim to limit the volume of congestion caused to
a constant weight parameter. [MulTCP][WindowPropFair] are
examples of weighted proportionally fair transports designed for
cost-fair environments [Rate_fair_Dis]. In this case, the
transport requires a count (not a ratio) of dropped/marked bytes
in the bit-congestible case and of dropped/marked packets in the
packet congestible case.
6. The State of the Art
The original 1993 paper on RED [RED93] proposed two options for the
RED active queue management algorithm: packet mode and byte mode.
Packet mode measured the queue length in packets and dropped (or
marked) individual packets with a probability independent of their
size. Byte mode measured the queue length in bytes and marked an
individual packet with probability in proportion to its size
(relative to the maximum packet size). In the paper's outline of
further work, it was stated that no recommendation had been made on
whether the queue size should be measured in bytes or packets, but
noted that the difference could be significant.
When RED was recommended for general deployment in 1998 [RFC2309],
the two modes were mentioned implying the choice between them was a
question of performance, referring to a 1997 email [pktByteEmail] for
advice on tuning. This email clarified that there were in fact two
orthogonal choices: whether to measure queue length in bytes or
packets (Section 6.1 below) and whether the drop probability of an
individual packet should depend on its own size (Section 6.2 below).
6.1. Congestion Measurement: Status
The choice of which metric to use to measure queue length was left
open in RFC2309. It is now well understood that queues for bit-
congestible resources should be measured in bytes, and queues for
packet-congestible resources should be measured in packets (see
Section 4).
Where buffers are not configured or legacy buffers cannot be
configured to the above guideline, we don't have to make allowances
for such legacy in future protocol design. If a bit-congestible
buffer is measured in packets, the operator will have set the
thresholds mindful of a typical mix of packets sizes. Any AQM
algorithm on such a buffer will be oversensitive to high proportions
of small packets, e.g. a DoS attack, and undersensitive to high
proportions of large packets. But an operator can safely keep such a
legacy buffer because any undersensitivity during unusual traffic
mixes cannot lead to congestion collapse given the buffer will
eventually revert to tail drop, discarding proportionately more large
packets.
Some modern queue implementations give a choice for setting RED's
thresholds in byte-mode or packet-mode. This may merely be an
administrator-interface preference, not altering how the queue itself
is measured but on some hardware it does actually change the way it
measures its queue. Whether a resource is bit-congestible or packet-
congestible is a property of the resource, so an admin SHOULD NOT
ever need to, or be able to, configure the way a queue measures
itself.
We believe the question of whether to measure queues in bytes or
packets is fairly well understood these days. The only outstanding
issues concern how to measure congestion when the queue is bit
congestible but the resource is packet congestible or vice versa (see
Section 4). But there is no controversy over what should be done.
It's just you have to be an expert in probability to work out what
should be done and, even if you have, it's not always easy to find a
practical algorithm to implement it.
6.2. Congestion Coding: Status 3.2. Congestion Coding: Status
6.2.1. Network Bias when Encoding 3.2.1. Network Bias when Encoding
The previously mentioned email [pktByteEmail] referred to by The previously mentioned email [pktByteEmail] referred to by
[RFC2309] said that the choice over whether a packet's own size [RFC2309] gave advice we now disagree with. It said that drop
should affect its drop probability "depends on the dominant end-to- probability should depend on the size of the packet being considered
end congestion control mechanisms". [Section 2 argues against this for drop if the resource is bit-congestible, but not if it is packet-
approach, citing the excellent advice in RFC3246.] The referenced congestible, but advised that most scarce resources in the Internet
email went on to argue that drop probability should depend on the were currently bit-congestible. The argument continued that if
size of the packet being considered for drop if the resource is bit- packet drops were inflated by packet size (byte-mode dropping), "a
congestible, but not if it is packet-congestible, but advised that flow's fraction of the packet drops is then a good indication of that
most scarce resources in the Internet were currently bit-congestible. flow's fraction of the link bandwidth in bits per second". This was
The argument continued that if packet drops were inflated by packet consistent with a referenced policing mechanism being worked on at
size (byte-mode dropping), "a flow's fraction of the packet drops is the time for detecting unusually high bandwidth flows, eventually
then a good indication of that flow's fraction of the link bandwidth published in 1999 [pBox]. [The problem could and should have been
in bits per second". This was consistent with a referenced policing solved by making the policing mechanism count the volume of bytes
mechanism being worked on at the time for detecting unusually high randomly dropped, not the number of packets.]
bandwidth flows, eventually published in 1999 [pBox]. [The problem
could have been solved by making the policing mechanism count the
volume of bytes randomly dropped, not the number of packets.]
A few months before RFC2309 was published, an addendum was added to A few months before RFC2309 was published, an addendum was added to
the above archived email referenced from the RFC, in which the final the above archived email referenced from the RFC, in which the final
paragraph seemed to partially retract what had previously been said. paragraph seemed to partially retract what had previously been said.
It clarified that the question of whether the probability of It clarified that the question of whether the probability of
dropping/marking a packet should depend on its size was not related dropping/marking a packet should depend on its size was not related
to whether the resource itself was bit congestible, but a completely to whether the resource itself was bit congestible, but a completely
orthogonal question. However the only example given had the queue orthogonal question. However the only example given had the queue
measured in packets but packet drop depended on the byte-size of the measured in packets but packet drop depended on the byte-size of the
packet in question. No example was given the other way round. packet in question. No example was given the other way round.
In 2000, Cnodder et al [REDbyte] pointed out that there was an error In 2000, Cnodder et al [REDbyte] pointed out that there was an error
in the part of the original 1993 RED algorithm that aimed to in the part of the original 1993 RED algorithm that aimed to
distribute drops uniformly, because it didn't correctly take into distribute drops uniformly, because it didn't correctly take into
account the adjustment for packet size. They recommended an account the adjustment for packet size. They recommended an
algorithm called RED_4 to fix this. But they also recommended a algorithm called RED_4 to fix this. But they also recommended a
further change, RED_5, to adjust drop rate dependent on the square of further change, RED_5, to adjust drop rate dependent on the square of
relative packet size. This was indeed consistent with one stated relative packet size. This was indeed consistent with one implied
motivation behind RED's byte mode drop--that we should reverse motivation behind RED's byte mode drop--that we should reverse
engineer the network to improve the performance of dominant end-to- engineer the network to improve the performance of dominant end-to-
end congestion control mechanisms. end congestion control mechanisms.
By 2003, a further change had been made to the adjustment for packet By 2003, a further change had been made to the adjustment for packet
size, this time in the RED algorithm of the ns2 simulator. Instead size, this time in the RED algorithm of the ns2 simulator. Instead
of taking each packet's size relative to a `maximum packet size' it of taking each packet's size relative to a `maximum packet size' it
was taken relative to a `mean packet size', intended to be a static was taken relative to a `mean packet size', intended to be a static
value representative of the `typical' packet size on the link. We value representative of the `typical' packet size on the link. We
have not been able to find a justification for this change in the have not been able to find a justification for this change in the
skipping to change at page 19, line 31 skipping to change at page 15, line 51
ring alarm bells hinting that there's a mistake in the theory ring alarm bells hinting that there's a mistake in the theory
somewhere]. On 10-Nov-2004, this variant of byte-mode packet drop somewhere]. On 10-Nov-2004, this variant of byte-mode packet drop
was made the default in the ns2 simulator. was made the default in the ns2 simulator.
The byte-mode drop variant of RED is, of course, not the only The byte-mode drop variant of RED is, of course, not the only
possible bias towards small packets in queueing algorithms. We have possible bias towards small packets in queueing algorithms. We have
already mentioned that tail-drop queues naturally tend to lock-out already mentioned that tail-drop queues naturally tend to lock-out
large packets once they are full. But also queues with fixed sized large packets once they are full. But also queues with fixed sized
buffers reduce the probability that small packets will be dropped if buffers reduce the probability that small packets will be dropped if
(and only if) they allow small packets to borrow buffers from the (and only if) they allow small packets to borrow buffers from the
pools for larger packets. As was explained in Section 4.1.1 on fixed pools for larger packets. As was explained in Section 3.1.1 on fixed
size buffer carving, borrowing effectively makes the maximum queue size buffer carving, borrowing effectively makes the maximum queue
size for small packets greater than that for large packets, because size for small packets greater than that for large packets, because
more buffers can be used by small packets while less will fit large more buffers can be used by small packets while less will fit large
packets. packets.
However, in itself, the bias towards small packets caused by buffer In itself, the bias towards small packets caused by buffer borrowing
borrowing is perfectly correct. Lower drop probability for small is perfectly correct. Lower drop probability for small packets is
packets is legitimate in buffer borrowing schemes, because small legitimate in buffer borrowing schemes, because small packets
packets genuinely congest the machine's buffer memory less than large genuinely congest the machine's buffer memory less than large
packets, given they can fit in more spaces. The bias towards small packets, given they can fit in more spaces. The bias towards small
packets is not artificially added (as it is in RED's byte-mode drop packets is not artificially added (as it is in RED's byte-mode drop
algorithm), it merely reflects the reality of the way fixed buffer algorithm), it merely reflects the reality of the way fixed buffer
memory gets congested. Incidentally, the bias towards small packets memory gets congested. Incidentally, the bias towards small packets
from buffer borrowing is nothing like as large as that of RED's byte- from buffer borrowing is nothing like as large as that of RED's byte-
mode drop. mode drop.
Nonetheless, fixed-buffer memory with tail drop is still prone to Nonetheless, fixed-buffer memory with tail drop is still prone to
lock-out large packets, purely because of the tail-drop aspect. So a lock-out large packets, purely because of the tail-drop aspect. So a
good AQM algorithm like RED with packet-mode drop should be used with good AQM algorithm like RED with packet-mode drop should be used with
fixed buffer memories where possible. If RED is too complicated to fixed buffer memories where possible. If RED is too complicated to
implement with multiple fixed buffer pools, the minimum necessary to implement with multiple fixed buffer pools, the minimum necessary to
prevent large packet lock-out is to ensure smaller packets never use prevent large packet lock-out is to ensure smaller packets never use
the last available buffer in any of the pools for larger packets. the last available buffer in any of the pools for larger packets.
6.2.2. Transport Bias when Decoding 3.2.2. Transport Bias when Decoding
The above proposals to alter the network layer to give a bias towards The above proposals to alter the network equipment to bias towards
smaller packets have largely carried on outside the IETF process smaller packets have largely carried on outside the IETF process
(unless one counts a reference in an informational RFC to an archived (unless one counts a reference in an informational RFC to an archived
email!). Whereas, within the IETF, there are many different email!). Whereas, within the IETF, there are many different
proposals to alter transport protocols to achieve the same goals, proposals to alter transport protocols to achieve the same goals,
i.e. either to make the flow bit-rate take account of packet size, or i.e. either to make the flow bit-rate take account of packet size, or
to protect control packets from loss. This memo argues that altering to protect control packets from loss. This memo argues that altering
transport protocols is the more principled approach. transport protocols is the more principled approach.
A recently approved experimental RFC adapts its transport layer A recently approved experimental RFC adapts its transport layer
protocol to take account of packet sizes relative to typical TCP protocol to take account of packet sizes relative to typical TCP
skipping to change at page 21, line 7 skipping to change at page 17, line 27
conclusive, instead reporting simulations of many of the conclusive, instead reporting simulations of many of the
possibilities in order to assess performance but not recommending any possibilities in order to assess performance but not recommending any
particular course of action. particular course of action.
The paper originally proposing TFRC with virtual packets (VP-TFRC) The paper originally proposing TFRC with virtual packets (VP-TFRC)
[CCvarPktSize] proposed that there should perhaps be two variants to [CCvarPktSize] proposed that there should perhaps be two variants to
cater for the different variants of RED. However, as the TFRC-SP cater for the different variants of RED. However, as the TFRC-SP
authors point out, there is no way for a transport to know whether authors point out, there is no way for a transport to know whether
some queues on its path have deployed RED with byte-mode packet drop some queues on its path have deployed RED with byte-mode packet drop
(except if an exhaustive survey found that no-one has deployed it!-- (except if an exhaustive survey found that no-one has deployed it!--
see Section 6.2.4). Incidentally, VP-TFRC also proposed that byte- see Section 3.2.4). Incidentally, VP-TFRC also proposed that byte-
mode RED dropping should really square the packet size compensation mode RED dropping should really square the packet size compensation
factor (like that of RED_5, but apparently unaware of it). factor (like that of RED_5, but apparently unaware of it).
Pre-congestion notification [I-D.ietf-pcn-marking-behaviour] is a Pre-congestion notification [I-D.ietf-pcn] is a proposal to use a
proposal to use a virtual queue for AQM marking for packets within virtual queue for AQM marking for packets within one Diffserv class
one Diffserv class in order to give early warning prior to any real in order to give early warning prior to any real queuing. The
queuing. The proposed PCN marking algorithms have been designed not proposed PCN marking algorithms have been designed not to take
to take account of packet size when forwarding through queues. account of packet size when forwarding through queues. Instead the
Instead the general principle has been to take account of the sizes general principle has been to take account of the sizes of marked
of marked packets when monitoring the fraction of marking at the edge packets when monitoring the fraction of marking at the edge of the
of the network. network.
6.2.3. Making Transports Robust against Control Packet Losses 3.2.3. Making Transports Robust against Control Packet Losses
Recently, two drafts have proposed changes to TCP that make it more Recently, two RFCs have defined changes to TCP that make it more
robust against losing small control packets [I-D.ietf-tcpm-ecnsyn] robust against losing small control packets [RFC5562] [RFC5690]. In
[I-D.floyd-tcpm-ackcc]. In both cases they note that the case for both cases they note that the case for these TCP changes would be
these TCP changes would be weaker if RED were biased against dropping weaker if RED were biased against dropping small packets. We argue
small packets. We argue here that these two proposals are a safer here that these two proposals are a safer and more principled way to
and more principled way to achieve TCP performance improvements than achieve TCP performance improvements than reverse engineering RED to
reverse engineering RED to benefit TCP. benefit TCP.
Although no proposals exist as far as we know, it would also be Although no proposals exist as far as we know, it would also be
possible and perfectly valid to make control packets robust against possible and perfectly valid to make control packets robust against
drop by explicitly requesting a lower drop probability using their drop by explicitly requesting a lower drop probability using their
Diffserv code point [RFC2474] to request a scheduling class with Diffserv code point [RFC2474] to request a scheduling class with
lower drop. lower drop.
The re-ECN protocol proposal [I-D.briscoe-tsvwg-re-ecn-tcp] is The re-ECN protocol proposal [I-D.briscoe-tsvwg-re-ecn-tcp] is
designed so that transports can be made more robust against losing designed so that transports can be made more robust against losing
control packets. It gives queues an incentive to optionally give control packets. It gives queues an incentive to optionally give
skipping to change at page 22, line 11 skipping to change at page 18, line 31
this would greatly improve the chances of short flows completing this would greatly improve the chances of short flows completing
quickly, but it would hardly increase traffic levels on the Internet, quickly, but it would hardly increase traffic levels on the Internet,
because Internet bytes have always been concentrated in the large because Internet bytes have always been concentrated in the large
flows. It further shows that the performance of many typical flows. It further shows that the performance of many typical
applications depends on completion of long serial chains of short applications depends on completion of long serial chains of short
messages. It argues that, given most of the value people get from messages. It argues that, given most of the value people get from
the Internet is concentrated within short flows, this simple the Internet is concentrated within short flows, this simple
expedient would greatly increase the value of the best efforts expedient would greatly increase the value of the best efforts
Internet at minimal cost. Internet at minimal cost.
6.2.4. Congestion Coding: Summary of Status 3.2.4. Congestion Coding: Summary of Status
+-----------+----------------+-----------------+--------------------+ +-----------+----------------+-----------------+--------------------+
| transport | RED_1 (packet | RED_4 (linear | RED_5 (square byte | | transport | RED_1 (packet | RED_4 (linear | RED_5 (square byte |
| cc | mode drop) | byte mode drop) | mode drop) | | cc | mode drop) | byte mode drop) | mode drop) |
+-----------+----------------+-----------------+--------------------+ +-----------+----------------+-----------------+--------------------+
| TCP or | s/sqrt(p) | sqrt(s/p) | 1/sqrt(p) | | TCP or | s/sqrt(p) | sqrt(s/p) | 1/sqrt(p) |
| TFRC | | | | | TFRC | | | |
| TFRC-SP | 1/sqrt(p) | 1/sqrt(sp) | 1/(s.sqrt(p)) | | TFRC-SP | 1/sqrt(p) | 1/sqrt(sp) | 1/(s.sqrt(p)) |
+-----------+----------------+-----------------+--------------------+ +-----------+----------------+-----------------+--------------------+
skipping to change at page 23, line 22 skipping to change at page 19, line 42
Turning to our more formal survey (Table 2), about 19% of those Turning to our more formal survey (Table 2), about 19% of those
surveyed have replied so far, giving a sample size of 16. Although surveyed have replied so far, giving a sample size of 16. Although
we do not have permission to identify the respondents, we can say we do not have permission to identify the respondents, we can say
that those that have responded include most of the larger vendors, that those that have responded include most of the larger vendors,
covering a large fraction of the market. They range across the large covering a large fraction of the market. They range across the large
network equipment vendors at L3 & L2, firewall vendors, wireless network equipment vendors at L3 & L2, firewall vendors, wireless
equipment vendors, as well as large software businesses with a small equipment vendors, as well as large software businesses with a small
selection of networking products. So far, all those who have selection of networking products. So far, all those who have
responded have confirmed that they have not implemented the variant responded have confirmed that they have not implemented the variant
of RED with drop dependent on packet size (2 are fairly sure they of RED with drop dependent on packet size (2 were fairly sure they
haven't but need to check more thoroughly). had not but needed to check more thoroughly).
+-------------------------------+----------------+-----------------+ +-------------------------------+----------------+-----------------+
| Response | No. of vendors | %age of vendors | | Response | No. of vendors | %age of vendors |
+-------------------------------+----------------+-----------------+ +-------------------------------+----------------+-----------------+
| Not implemented | 14 | 17% | | Not implemented | 14 | 17% |
| Not implemented (probably) | 2 | 2% | | Not implemented (probably) | 2 | 2% |
| Implemented | 0 | 0% | | Implemented | 0 | 0% |
| No response | 68 | 81% | | No response | 68 | 81% |
| Total companies/orgs surveyed | 84 | 100% | | Total companies/orgs surveyed | 84 | 100% |
+-------------------------------+----------------+-----------------+ +-------------------------------+----------------+-----------------+
skipping to change at page 23, line 49 skipping to change at page 20, line 29
code has been most prevalent, though one vendor had a more principled code has been most prevalent, though one vendor had a more principled
reason for avoiding it--similar to the argument of this document. We reason for avoiding it--similar to the argument of this document. We
have established that Linux does not implement RED with packet size have established that Linux does not implement RED with packet size
drop bias, although we have not investigated a wider range of open drop bias, although we have not investigated a wider range of open
source code. source code.
Finally, we repeat that RED's byte mode drop is not the only way to Finally, we repeat that RED's byte mode drop is not the only way to
bias towards small packets--tail-drop tends to lock-out large packets bias towards small packets--tail-drop tends to lock-out large packets
very effectively. Our survey was of vendor implementations, so we very effectively. Our survey was of vendor implementations, so we
cannot be certain about operator deployment. But we believe many cannot be certain about operator deployment. But we believe many
queues in the Internet are still tail-drop. My own company (BT) has queues in the Internet are still tail-drop. The company of one of
widely deployed RED, but there are bound to be many tail-drop queues, the co-authors (BT) has widely deployed RED, but there are bound to
particularly in access network equipment and on middleboxes like be many tail-drop queues, particularly in access network equipment
firewalls, where RED is not always available. Routers using a memory and on middleboxes like firewalls, where RED is not always available.
architecture based on fixed size buffers with borrowing may also
still be prevalent in the Internet. As explained in Section 6.2.1,
these also provide a marginal (but legitimate) bias towards small
packets. So even though RED byte-mode drop is not prevalent, it is
likely there is still some bias towards small packets in the Internet
due to tail drop and fixed buffer borrowing.
7. Outstanding Issues and Next Steps Routers using a memory architecture based on fixed size buffers with
borrowing may also still be prevalent in the Internet. As explained
in Section 3.2.1, these also provide a marginal (but legitimate) bias
towards small packets. So even though RED byte-mode drop is not
prevalent, it is likely there is still some bias towards small
packets in the Internet due to tail drop and fixed buffer borrowing.
7.1. Bit-congestible World 4. Outstanding Issues and Next Steps
4.1. Bit-congestible World
For a connectionless network with nearly all resources being bit- For a connectionless network with nearly all resources being bit-
congestible we believe the recommended position is now unarguably congestible we believe the recommended position is now unarguably
clear--that the network should not make allowance for packet sizes clear--that the network should not make allowance for packet sizes
and the transport should. This leaves two outstanding issues: and the transport should. This leaves two outstanding issues:
o How to handle any legacy of AQM with byte-mode drop already o How to handle any legacy of AQM with byte-mode drop already
deployed; deployed;
o The need to start a programme to update transport congestion o The need to start a programme to update transport congestion
control protocol standards to take account of packet size. control protocol standards to take account of packet size.
The sample of returns from our vendor survey Section 6.2.4 suggest The sample of returns from our vendor survey Section 3.2.4 suggest
that byte-mode packet drop seems not to be implemented at all let that byte-mode packet drop seems not to be implemented at all let
alone deployed, or if it is, it is likely to be very sparse. alone deployed, or if it is, it is likely to be very sparse.
Therefore, we do not really need a migration strategy from all but Therefore, we do not really need a migration strategy from all but
nothing to nothing. nothing to nothing.
A programme of standards updates to take account of packet size in A programme of standards updates to take account of packet size in
transport congestion control protocols has started with TFRC-SP transport congestion control protocols has started with TFRC-SP
[RFC4828], while weighted TCPs implemented in the research community [RFC4828], while weighted TCPs implemented in the research community
[WindowPropFair] could form the basis of a future change to TCP [WindowPropFair] could form the basis of a future change to TCP
congestion control [RFC5681] itself. congestion control [RFC5681] itself.
7.2. Bit- & Packet-congestible World 4.2. Bit- & Packet-congestible World
Nonetheless, a connectionless network with both bit-congestible and Nonetheless, a connectionless network with both bit-congestible and
packet-congestible resources is a different matter. If we believe we packet-congestible resources is a different matter. If we believe we
should allow for this possibility in the future, this space contains should allow for this possibility in the future, this space contains
a truly open research issue. a truly open research issue.
The idealised wire protocol coding described in Section 5 requires at We develop the concept of an idealised congestion notification
protocol that supports both bit-congestible and packet-congestible
resources in Appendix B. The congestion notification requires at
least two flags for congestion of bit-congestible and packet- least two flags for congestion of bit-congestible and packet-
congestible resources. This hides a fundamental problem--much more congestible resources. This hides a fundamental problem--much more
fundamental than whether we can magically create header space for yet fundamental than whether we can magically create header space for yet
another ECN flag in IPv4, or whether it would work while being another ECN flag in IPv4, or whether it would work while being
deployed incrementally. A congestion notification protocol must deployed incrementally. A congestion notification protocol must
survive a transition from low levels of congestion to high. Marking survive a transition from low levels of congestion to high. Marking
two states is feasible with explicit marking, but much harder if two states is feasible with explicit marking, but much harder if
packets are dropped. Also, it will not always be cost-effective to packets are dropped. Also, it will not always be cost-effective to
implement AQM at every low level resource, so drop will often have to implement AQM at every low level resource, so drop will often have to
suffice. Distinguishing drop from delivery naturally provides just suffice. Distinguishing drop from delivery naturally provides just
skipping to change at page 25, line 33 skipping to change at page 22, line 14
on the premise that, as network processors become more cost on the premise that, as network processors become more cost
effective, per packet operations will become more complex effective, per packet operations will become more complex
(irrespective of whether more function in the network layer is (irrespective of whether more function in the network layer is
desirable). Consequently the premise is that CPU congestion will desirable). Consequently the premise is that CPU congestion will
become more common. DRQ is a proposed modification to the RED become more common. DRQ is a proposed modification to the RED
algorithm that folds both bit congestion and packet congestion into algorithm that folds both bit congestion and packet congestion into
one signal (either loss or ECN). one signal (either loss or ECN).
The problem of signalling packet processing congestion is not The problem of signalling packet processing congestion is not
pressing, as most Internet resources are designed to be bit- pressing, as most Internet resources are designed to be bit-
congestible before packet processing starts to congest. However, the congestible before packet processing starts to congest (see
IRTF Internet congestion control research group (ICCRG) has set Section 1.1). However, the IRTF Internet congestion control research
itself the task of reaching consensus on generic forwarding group (ICCRG) has set itself the task of reaching consensus on
mechanisms that are necessary and sufficient to support the generic forwarding mechanisms that are necessary and sufficient to
Internet's future congestion control requirements (the first support the Internet's future congestion control requirements (the
challenge in first challenge in [I-D.irtf-iccrg-welzl]). Therefore, rather than
[I-D.irtf-iccrg-welzl-congestion-control-open-research]). Therefore, not giving this problem any thought at all, just because it is hard
rather than not giving this problem any thought at all, just because and currently hypothetical, we defer the question of whether packet
it is hard and currently hypothetical, we defer the question of congestion might become common and what to do if it does to the IRTF
whether packet congestion might become common and what to do if it (the 'Small Packets' challenge in [I-D.irtf-iccrg-welzl]).
does to the IRTF (the 'Small Packets' challenge in
[I-D.irtf-iccrg-welzl-congestion-control-open-research]).
8. Security Considerations 5. Recommendation and Conclusions
This draft recommends that queues do not bias drop probability 5.1. Recommendation on Queue Measurement
towards small packets as this creates a perverse incentive for
transports to break down their flows into tiny segments. One of the
benefits of implementing AQM was meant to be to remove this perverse
incentive that drop-tail queues gave to small packets. Of course, if
transports really want to make the greatest gains, they don't have to
respond to congestion anyway. But we don't want applications that
are trying to behave to discover that they can go faster by using
smaller packets.
In practice, transports cannot all be trusted to respond to Queue length is usually the most correct and simplest way to measure
congestion. So another reason for recommending that queues do not congestion of a resource. To avoid the pathological effects of drop
bias drop probability towards small packets is to avoid the tail, an AQM function can then be used to transform queue length into
vulnerability to small packet DDoS attacks that would otherwise the probability of dropping or marking a packet (e.g. RED's
result. One of the benefits of implementing AQM was meant to be to piecewise linear function between thresholds).
remove drop-tail's DoS vulnerability to small packets, so we
shouldn't add it back again.
If most queues implemented AQM with byte-mode drop, the resulting If the resource is bit-congestible, the length of the queue SHOULD be
network would amplify the potency of a small packet DDoS attack. At measured in bytes. If the resource is packet-congestible, the length
the first queue the stream of packets would push aside a greater of the queue SHOULD be measured in packets. No other choice makes
proportion of large packets, so more of the small packets would sense, because the number of packets waiting in the queue isn't
survive to attack the next queue. Thus a flood of small packets relevant if the resource gets congested by bytes and vice versa. We
would continue on towards the destination, pushing regular traffic discuss the implications on RED's byte mode and packet mode for
with large packets out of the way in one queue after the next, but measuring queue length in Section 3.
suffering much less drop itself.
Appendix C explains why the ability of networks to police the NOTE WELL that RED's byte-mode queue measurement is fine, being
response of _any_ transport to congestion depends on bit-congestible completely orthogonal to byte-mode drop. If a RED implementation has
network resources only doing packet-mode not byte-mode drop. In a byte-mode but does not specify what sort of byte-mode, it is most
summary, it says that making drop probability depend on the size of probably byte-mode queue measurement, which is fine. However, if in
the packets that bits happen to be divided into simply encourages the doubt, the vendor should be consulted.
bits to be divided into smaller packets. Byte-mode drop would
therefore irreversibly complicate any attempt to fix the Internet's
incentive structures.
9. Conclusions 5.2. Recommendation on Notifying Congestion
The strong conclusion is that AQM algorithms such as RED SHOULD NOT The strong recommendation is that AQM algorithms such as RED SHOULD
use byte-mode drop. More generally, the Internet's congestion NOT use byte-mode drop. More generally, the Internet's congestion
notification protocols (drop, ECN & PCN) SHOULD take account of notification protocols (drop, ECN & PCN) SHOULD take account of
packet size when the notification is read by the transport layer, NOT packet size when the notification is read by the transport layer, NOT
when it is written by the network layer. This approach offers when it is written by the network layer. This approach offers
sufficient and correct congestion information for all known and sufficient and correct congestion information for all known and
future transport protocols and also ensures no perverse incentives future transport protocols and also ensures no perverse incentives
are created that would encourage transports to use inappropriately are created that would encourage transports to use inappropriately
small packet sizes. small packet sizes.
The alternative of deflating RED's drop probability for smaller The alternative of deflating RED's drop probability for smaller
packet sizes (byte-mode drop) has no enduring advantages. It is more packet sizes (byte-mode drop) has no enduring advantages. It is more
complex, it creates the perverse incentive to fragment segments into complex, it creates the perverse incentive to fragment segments into
tiny pieces and it reopens the vulnerability to floods of small- tiny pieces and it reopens the vulnerability to floods of small-
packets that drop-tail queues suffered from and AQM was designed to packets that drop-tail queues suffered from and AQM was designed to
remove. Byte-mode drop is a change to the network layer that makes remove.
allowance for an omission from the design of TCP, effectively reverse
Byte-mode drop is a change to the network layer that makes allowance
for an omission from the design of TCP, effectively reverse
engineering the network layer to contrive to make two TCPs with engineering the network layer to contrive to make two TCPs with
different packet sizes run at equal bit rates (rather than packet different packet sizes run at equal bit rates (rather than packet
rates) under the same path conditions. It also improves TCP rates) under the same path conditions.
performance by reducing the chance that a SYN or a pure ACK will be
dropped, because they are small. But we SHOULD NOT hack the network It also improves TCP performance by reducing the chance that a SYN or
layer to improve or fix certain transport protocols. No matter how a pure ACK will be dropped, because they are small. But we SHOULD
predominant a transport protocol is (even if it's TCP), trying to NOT hack the network layer to improve or fix certain transport
correct for its failings by biasing towards small packets in the protocols. No matter how predominant a transport protocol is (even
network layer creates a perverse incentive to break down all flows if it's TCP), trying to correct for its failings by biasing towards
from all transports into tiny segments. small packets in the network layer creates a perverse incentive to
break down all flows from all transports into tiny segments.
So far, our survey of 84 vendors across the industry has drawn So far, our survey of 84 vendors across the industry has drawn
responses from about 19%, none of whom have implemented the byte mode responses from about 19%, none of whom have implemented the byte mode
packet drop variant of RED. Given there appears to be little, if packet drop variant of RED. Given there appears to be little, if
any, installed base it seems we can recommend removal of byte-mode any, installed base it seems we can recommend removal of byte-mode
drop from RED with little, if any, incremental deployment impact. drop from RED with little, if any, incremental deployment impact.
If a vendor has implemented byte-mode drop, and an operator has If a vendor has implemented byte-mode drop, and an operator has
turned it on, it is strongly RECOMMENDED that it SHOULD be turned turned it on, it is strongly RECOMMENDED that it SHOULD be turned
off. Note that RED as a whole SHOULD NOT be turned off, as without off. Note that RED as a whole SHOULD NOT be turned off, as without
it, a drop tail queue also biases against large packets. But note it, a drop tail queue also biases against large packets. But note
also that turning off byte-mode may alter the relative performance of also that turning off byte-mode may alter the relative performance of
applications using different packet sizes, so it would be advisable applications using different packet sizes, so it would be advisable
to establish the implications before turning it off. to establish the implications before turning it off.
Instead, the IETF transport area should continue its programme of 5.3. Recommendation on Responding to Congestion
updating congestion control protocols to take account of packet size
and to make transports less sensitive to losing control packets like
SYNs and pure ACKS.
NOTE WELL that RED's byte-mode queue measurement is fine, being Instead of network equipment biasing its congestion notification for
completely orthogonal to byte-mode drop. If a RED implementation has small packets, the IETF transport area should continue its programme
a byte-mode but does not specify what sort of byte-mode, it is most of updating congestion control protocols to take account of packet
probably byte-mode queue measurement, which is fine. However, if in size and to make transports less sensitive to losing control packets
doubt, the vendor should be consulted. like SYNs and pure ACKS.
5.4. Recommended Future Research
The above conclusions cater for the Internet as it is today with The above conclusions cater for the Internet as it is today with
most, if not all, resources being primarily bit-congestible. A most, if not all, resources being primarily bit-congestible. A
secondary conclusion of this memo is that we may see more packet- secondary conclusion of this memo is that we may see more packet-
congestible resources in the future, so research may be needed to congestible resources in the future, so research may be needed to
extend the Internet's congestion notification (drop or ECN) so that extend the Internet's congestion notification (drop or ECN) so that
it can handle a mix of bit-congestible and packet-congestible it can handle a mix of bit-congestible and packet-congestible
resources. resources.
10. Acknowledgements 6. Security Considerations
This draft recommends that queues do not bias drop probability
towards small packets as this creates a perverse incentive for
transports to break down their flows into tiny segments. One of the
benefits of implementing AQM was meant to be to remove this perverse
incentive that drop-tail queues gave to small packets. Of course, if
transports really want to make the greatest gains, they don't have to
respond to congestion anyway. But we don't want applications that
are trying to behave to discover that they can go faster by using
smaller packets.
In practice, transports cannot all be trusted to respond to
congestion. So another reason for recommending that queues do not
bias drop probability towards small packets is to avoid the
vulnerability to small packet DDoS attacks that would otherwise
result. One of the benefits of implementing AQM was meant to be to
remove drop-tail's DoS vulnerability to small packets, so we
shouldn't add it back again.
If most queues implemented AQM with byte-mode drop, the resulting
network would amplify the potency of a small packet DDoS attack. At
the first queue the stream of packets would push aside a greater
proportion of large packets, so more of the small packets would
survive to attack the next queue. Thus a flood of small packets
would continue on towards the destination, pushing regular traffic
with large packets out of the way in one queue after the next, but
suffering much less drop itself.
Appendix C explains why the ability of networks to police the
response of _any_ transport to congestion depends on bit-congestible
network resources only doing packet-mode not byte-mode drop. In
summary, it says that making drop probability depend on the size of
the packets that bits happen to be divided into simply encourages the
bits to be divided into smaller packets. Byte-mode drop would
therefore irreversibly complicate any attempt to fix the Internet's
incentive structures.
7. Acknowledgements
Thank you to Sally Floyd, who gave extensive and useful review Thank you to Sally Floyd, who gave extensive and useful review
comments. Also thanks for the reviews from Philip Eardley, Toby comments. Also thanks for the reviews from Philip Eardley, Toby
Moncaster and Arnaud Jacquet as well as helpful explanations of Moncaster and Arnaud Jacquet as well as helpful explanations of
different hardware approaches from Larry Dunn and Fred Baker. I am different hardware approaches from Larry Dunn and Fred Baker. I am
grateful to Bruce Davie and his colleagues for providing a timely and grateful to Bruce Davie and his colleagues for providing a timely and
efficient survey of RED implementation in Cisco's product range. efficient survey of RED implementation in Cisco's product range.
Also grateful thanks to Toby Moncaster, Will Dormann, John Regnault, Also grateful thanks to Toby Moncaster, Will Dormann, John Regnault,
Simon Carter and Stefaan De Cnodder who further helped survey the Simon Carter and Stefaan De Cnodder who further helped survey the
current status of RED implementation and deployment and, finally, current status of RED implementation and deployment and, finally,
thanks to the anonymous individuals who responded. thanks to the anonymous individuals who responded.
Bob Briscoe is partly funded by Trilogy, a research project (ICT- Bob Briscoe and Jukka Manner are partly funded by Trilogy, a research
216372) supported by the European Community under its Seventh project (ICT- 216372) supported by the European Community under its
Framework Programme. The views expressed here are those of the Seventh Framework Programme. The views expressed here are those of
author only. the authors only.
11. Comments Solicited 8. Comments Solicited
Comments and questions are encouraged and very welcome. They can be Comments and questions are encouraged and very welcome. They can be
addressed to the IETF Transport Area working group mailing list addressed to the IETF Transport Area working group mailing list
<tsvwg@ietf.org>, and/or to the authors. <tsvwg@ietf.org>, and/or to the authors.
12. References 9. References
12.1. Normative References 9.1. Normative References
[RFC2119] Bradner, S., "Key words for use in RFCs to Indicate [RFC2119] Bradner, S., "Key words for use in
Requirement Levels", BCP 14, RFC 2119, March 1997. RFCs to Indicate Requirement Levels",
BCP 14, RFC 2119, March 1997.
[RFC2309] Braden, B., Clark, D., Crowcroft, J., Davie, B., Deering, [RFC2309] Braden, B., Clark, D., Crowcroft, J.,
S., Estrin, D., Floyd, S., Jacobson, V., Minshall, G., Davie, B., Deering, S., Estrin, D.,
Partridge, C., Peterson, L., Ramakrishnan, K., Shenker, Floyd, S., Jacobson, V., Minshall,
S., Wroclawski, J., and L. Zhang, "Recommendations on G., Partridge, C., Peterson, L.,
Queue Management and Congestion Avoidance in the Ramakrishnan, K., Shenker, S.,
Internet", RFC 2309, April 1998. Wroclawski, J., and L. Zhang,
"Recommendations on Queue Management
and Congestion Avoidance in the
Internet", RFC 2309, April 1998.
[RFC3168] Ramakrishnan, K., Floyd, S., and D. Black, "The Addition [RFC3168] Ramakrishnan, K., Floyd, S., and D.
of Explicit Congestion Notification (ECN) to IP", Black, "The Addition of Explicit
RFC 3168, September 2001. Congestion Notification (ECN) to IP",
RFC 3168, September 2001.
[RFC3426] Floyd, S., "General Architectural and Policy [RFC3426] Floyd, S., "General Architectural and
Considerations", RFC 3426, November 2002. Policy Considerations", RFC 3426,
November 2002.
[RFC5033] Floyd, S. and M. Allman, "Specifying New Congestion [RFC5033] Floyd, S. and M. Allman, "Specifying
Control Algorithms", BCP 133, RFC 5033, August 2007. New Congestion Control Algorithms",
BCP 133, RFC 5033, August 2007.
12.2. Informative References 9.2. Informative References
[CCvarPktSize] [CCvarPktSize] Widmer, J., Boutremans, C., and J-Y.
Widmer, J., Boutremans, C., and J-Y. Le Boudec, Le Boudec, "Congestion Control for
"Congestion Control for Flows with Variable Packet Size", Flows with Variable Packet Size", ACM
ACM CCR 34(2) 137--151, 2004, CCR 34(2) 137--151, 2004, <http://
<http://doi.acm.org/10.1145/997150.997162>. doi.acm.org/10.1145/997150.997162>.
[DRQ] Shin, M., Chong, S., and I. Rhee, "Dual-Resource TCP/AQM [DRQ] Shin, M., Chong, S., and I. Rhee,
for Processing-Constrained Networks", IEEE/ACM "Dual-Resource TCP/AQM for
Transactions on Networking Vol 16, issue 2, April 2008, Processing-Constrained Networks",
<http://dx.doi.org/10.1109/TNET.2007.900415>. IEEE/ACM Transactions on
Networking Vol 16, issue 2,
April 2008, <http://dx.doi.org/
10.1109/TNET.2007.900415>.
[DupTCP] Wischik, D., "Short messages", Royal Society workshop on [DupTCP] Wischik, D., "Short messages", Royal
networks: modelling and control , September 2007, <http:// Society workshop on networks:
www.cs.ucl.ac.uk/staff/ucacdjw/Research/shortmsg.html>. modelling and control ,
September 2007, <http://
www.cs.ucl.ac.uk/staff/ucacdjw/
Research/shortmsg.html>.
[ECNFixedWireless] [ECNFixedWireless] Siris, V., "Resource Control for
Siris, V., "Resource Control for Elastic Traffic in CDMA Elastic Traffic in CDMA Networks",
Networks", Proc. ACM MOBICOM'02 , September 2002, <http:// Proc. ACM MOBICOM'02 ,
www.ics.forth.gr/netlab/publications/ September 2002, <http://
resource_control_elastic_cdma.html>. www.ics.forth.gr/netlab/publications/
resource_control_elastic_cdma.html>.
[Evol_cc] Gibbens, R. and F. Kelly, "Resource pricing and the [Evol_cc] Gibbens, R. and F. Kelly, "Resource
evolution of congestion control", Automatica 35(12)1969-- pricing and the evolution of
1985, December 1999, congestion control",
<http://www.statslab.cam.ac.uk/~frank/evol.html>. Automatica 35(12)1969--1985,
December 1999, <http://
www.statslab.cam.ac.uk/~frank/
evol.html>.
[I-D.briscoe-tsvwg-re-ecn-tcp] [I-D.briscoe-tsvwg-re-ecn-tcp] Briscoe, B., Jacquet, A., Moncaster,
Briscoe, B., Jacquet, A., Moncaster, T., and A. Smith, T., and A. Smith, "Re-ECN: Adding
"Re-ECN: Adding Accountability for Causing Congestion to Accountability for Causing Congestion
TCP/IP", draft-briscoe-tsvwg-re-ecn-tcp-07 (work in to TCP/IP",
progress), March 2009. draft-briscoe-tsvwg-re-ecn-tcp-08
(work in progress), September 2009.
[I-D.floyd-tcpm-ackcc] [I-D.ietf-pcn] Eardley, P., "Metering and marking
Floyd, S., "Adding Acknowledgement Congestion Control to behaviour of PCN-nodes",
TCP", draft-floyd-tcpm-ackcc-06 (work in progress), draft-ietf-pcn-marking-behaviour-05
July 2009. (work in progress), August 2009.
[I-D.ietf-pcn-marking-behaviour] [I-D.irtf-iccrg-welzl] Welzl, M., Scharf, M., Briscoe, B.,
Eardley, P., "Metering and marking behaviour of PCN- and D. Papadimitriou, "Open Research
nodes", draft-ietf-pcn-marking-behaviour-05 (work in Issues in Internet Congestion
progress), August 2009. Control", draft-irtf-iccrg-welzl-
congestion-control-open-research-07
(work in progress), June 2010.
[I-D.ietf-tcpm-ecnsyn] [IOSArch] Bollapragada, V., White, R., and C.
Floyd, S., "Adding Explicit Congestion Notification (ECN) Murphy, "Inside Cisco IOS Software
Capability to TCP's SYN/ACK Packets", Architecture", Cisco Press: CCIE
draft-ietf-tcpm-ecnsyn-10 (work in progress), May 2009. Professional Development ISBN13: 978-
1-57870-181-0, July 2000.
[I-D.irtf-iccrg-welzl-congestion-control-open-research] [MulTCP] Crowcroft, J. and Ph. Oechslin,
Welzl, M., Scharf, M., Briscoe, B., and D. Papadimitriou, "Differentiated End to End Internet
"Open Research Issues in Internet Congestion Control", Services using a Weighted
draft-irtf-iccrg-welzl-congestion-control-open-research-05 Proportional Fair Sharing TCP",
(work in progress), September 2009. CCR 28(3) 53--69, July 1998, <http://
www.cs.ucl.ac.uk/staff/J.Crowcroft/
hipparch/pricing.html>.
[IOSArch] Bollapragada, V., White, R., and C. Murphy, "Inside Cisco [PktSizeEquCC] Vasallo, P., "Variable Packet Size
IOS Software Architecture", Cisco Press: CCIE Professional Equation-Based Congestion Control",
Development ISBN13: 978-1-57870-181-0, July 2000. ICSI Technical Report tr-00-008,
2000, <http://http.icsi.berkeley.edu/
ftp/global/pub/techreports/2000/
tr-00-008.pdf>.
[MulTCP] Crowcroft, J. and Ph. Oechslin, "Differentiated End to End [RED93] Floyd, S. and V. Jacobson, "Random
Internet Services using a Weighted Proportional Fair Early Detection (RED) gateways for
Sharing TCP", CCR 28(3) 53--69, July 1998, <http:// Congestion Avoidance", IEEE/ACM
www.cs.ucl.ac.uk/staff/J.Crowcroft/hipparch/pricing.html>. Transactions on Networking 1(4) 397--
413, August 1993, <http://
www.icir.org/floyd/papers/red/
red.html>.
[PktSizeEquCC] [REDbias] Eddy, W. and M. Allman, "A Comparison
Vasallo, P., "Variable Packet Size Equation-Based of RED's Byte and Packet Modes",
Congestion Control", ICSI Technical Report tr-00-008, Computer Networks 42(3) 261--280,
2000, <http://http.icsi.berkeley.edu/ftp/global/pub/ June 2003, <http://www.ir.bbn.com/
techreports/2000/tr-00-008.pdf>. documents/articles/redbias.ps>.
[RED93] Floyd, S. and V. Jacobson, "Random Early Detection (RED) [REDbyte] De Cnodder, S., Elloumi, O., and K.
gateways for Congestion Avoidance", IEEE/ACM Transactions Pauwels, "RED behavior with different
on Networking 1(4) 397--413, August 1993, packet sizes", Proc. 5th IEEE
<http://www.icir.org/floyd/papers/red/red.html>. Symposium on Computers and
Communications (ISCC) 793--799,
July 2000, <http://www.icir.org/
floyd/red/Elloumi99.pdf>.
[REDbias] Eddy, W. and M. Allman, "A Comparison of RED's Byte and [RFC2474] Nichols, K., Blake, S., Baker, F.,
Packet Modes", Computer Networks 42(3) 261--280, and D. Black, "Definition of the
June 2003, Differentiated Services Field (DS
<http://www.ir.bbn.com/documents/articles/redbias.ps>. Field) in the IPv4 and IPv6 Headers",
RFC 2474, December 1998.
[REDbyte] De Cnodder, S., Elloumi, O., and K. Pauwels, "RED behavior [RFC3448] Handley, M., Floyd, S., Padhye, J.,
with different packet sizes", Proc. 5th IEEE Symposium on and J. Widmer, "TCP Friendly Rate
Computers and Communications (ISCC) 793--799, July 2000, Control (TFRC): Protocol
<http://www.icir.org/floyd/red/Elloumi99.pdf>. Specification", RFC 3448,
January 2003.
[RFC2474] Nichols, K., Blake, S., Baker, F., and D. Black, [RFC3714] Floyd, S. and J. Kempf, "IAB Concerns
"Definition of the Differentiated Services Field (DS Regarding Congestion Control for
Field) in the IPv4 and IPv6 Headers", RFC 2474, Voice Traffic in the Internet",
December 1998. RFC 3714, March 2004.
[RFC3448] Handley, M., Floyd, S., Padhye, J., and J. Widmer, "TCP [RFC4782] Floyd, S., Allman, M., Jain, A., and
Friendly Rate Control (TFRC): Protocol Specification", P. Sarolahti, "Quick-Start for TCP
RFC 3448, January 2003. and IP", RFC 4782, January 2007.
[RFC3714] Floyd, S. and J. Kempf, "IAB Concerns Regarding Congestion [RFC4828] Floyd, S. and E. Kohler, "TCP
Control for Voice Traffic in the Internet", RFC 3714, Friendly Rate Control (TFRC): The
March 2004. Small-Packet (SP) Variant", RFC 4828,
April 2007.
[RFC4782] Floyd, S., Allman, M., Jain, A., and P. Sarolahti, "Quick- [RFC5562] Kuzmanovic, A., Mondal, A., Floyd,
Start for TCP and IP", RFC 4782, January 2007. S., and K. Ramakrishnan, "Adding
Explicit Congestion Notification
(ECN) Capability to TCP's SYN/ACK
Packets", RFC 5562, June 2009.
[RFC4828] Floyd, S. and E. Kohler, "TCP Friendly Rate Control [RFC5670] Eardley, P., "Metering and Marking
(TFRC): The Small-Packet (SP) Variant", RFC 4828, Behaviour of PCN-Nodes", RFC 5670,
April 2007. November 2009.
[RFC5681] Allman, M., Paxson, V., and E. Blanton, "TCP Congestion [RFC5681] Allman, M., Paxson, V., and E.
Control", RFC 5681, September 2009. Blanton, "TCP Congestion Control",
RFC 5681, September 2009.
[Rate_fair_Dis] [RFC5690] Floyd, S., Arcia, A., Ros, D., and J.
Briscoe, B., "Flow Rate Fairness: Dismantling a Religion", Iyengar, "Adding Acknowledgement
ACM CCR 37(2)63--74, April 2007, Congestion Control to TCP", RFC 5690,
<http://portal.acm.org/citation.cfm?id=1232926>. February 2010.
[WindowPropFair] [Rate_fair_Dis] Briscoe, B., "Flow Rate Fairness:
Siris, V., "Service Differentiation and Performance of Dismantling a Religion", ACM
Weighted Window-Based Congestion Control and Packet CCR 37(2)63--74, April 2007, <http://
Marking Algorithms in ECN Networks", Computer portal.acm.org/
Communications 26(4) 314--326, 2002, <http:// citation.cfm?id=1232926>.
www.ics.forth.gr/netgroup/publications/
weighted_window_control.html>.
[gentle_RED] [WindowPropFair] Siris, V., "Service Differentiation
Floyd, S., "Recommendation on using the "gentle_" variant and Performance of Weighted Window-
of RED", Web page , March 2000, Based Congestion Control and Packet
<http://www.icir.org/floyd/red/gentle.html>. Marking Algorithms in ECN Networks",
Computer Communications 26(4) 314--
326, 2002, <http://www.ics.forth.gr/
netgroup/publications/
weighted_window_control.html>.
[pBox] Floyd, S. and K. Fall, "Promoting the Use of End-to-End [gentle_RED] Floyd, S., "Recommendation on using
Congestion Control in the Internet", IEEE/ACM Transactions the "gentle_" variant of RED", Web
on Networking 7(4) 458--472, August 1999, page , March 2000, <http://
<http://www.aciri.org/floyd/end2end-paper.html>. www.icir.org/floyd/red/gentle.html>.
[pktByteEmail] [pBox] Floyd, S. and K. Fall, "Promoting the
Floyd, S., "RED: Discussions of Byte and Packet Modes", Use of End-to-End Congestion Control
email , March 1997, in the Internet", IEEE/ACM
<http://www-nrg.ee.lbl.gov/floyd/REDaveraging.txt>. Transactions on Networking 7(4) 458--
472, August 1999, <http://
www.aciri.org/floyd/
end2end-paper.html>.
[xcp-spec] [pktByteEmail] Yes and J. Doe, "Missing for now",
Falk, A., "Specification for the Explicit Control Protocol RFC 0000, May 2006.
(XCP)", draft-falk-xcp-spec-03 (work in progress),
July 2007.
(Expired) [xcp-spec] Falk, A., "Specification for the
Explicit Control Protocol (XCP)",
draft-falk-xcp-spec-03 (work in
progress), July 2007.
Editorial Comments Appendix A. Congestion Notification Definition: Further Justification
[Note_Variation] The algorithm of the byte-mode drop variant of RED In Section 1.1 on the definition of congestion notification, load not
switches off any bias towards small packets capacity was used as the denominator. This also has a subtle
whenever the smoothed queue length dictates that significance in the related debate over the design of new transport
the drop probability of large packets should be protocols--typical new protocol designs (e.g. in XCP [xcp-spec] &
100%. In the example in the Introduction, as the Quickstart [RFC4782]) expect the sending transport to communicate its
large packet drop probability varies around 25% the desired flow rate to the network and network elements to
small packet drop probability will vary around 1%, progressively subtract from this so that the achievable flow rate
but with occasional jumps to 100% whenever the emerges at the receiving transport.
instantaneous queue (after drop) manages to sustain
a length above the 100% drop point for longer than
the queue averaging period.
Appendix A. Example Scenarios Congestion notification with total load in the denominator can serve
a similar purpose (though in retrospect not in advance like XCP &
QuickStart). Congestion notification is a dimensionless fraction but
each source can extract necessary rate information from it because it
already knows what its own rate is. Even though congestion
notification doesn't communicate a rate explicitly, from each
source's point of view congestion notification represents the
fraction of the rate it was sending a round trip ago that couldn't
(or wouldn't) be served by available resources.
A.1. Notation Appendix B. Idealised Wire Protocol
To prove our idealised wire protocol (Section 5) is correct, we will We will start by inventing an idealised congestion notification
compare two flows with different packet sizes, s_1 and s_2 [bit/pkt], protocol before discussing how to make it practical. The idealised
to make sure their transports each see the correct congestion protocol is shown to be correct using examples later in this
appendix.
B.1. Protocol Coding
Congestion notification involves the congested resource coding a
congestion notification signal into the packet stream and the
transports decoding it. The idealised protocol uses two different
(imaginary) fields in each datagram to signal congestion: one for
byte congestion and one for packet congestion.
We are not saying two ECN fields will be needed (and we are not
saying that somehow a resource should be able to drop a packet in one
of two different ways so that the transport can distinguish which
sort of drop it was!). These two congestion notification channels
are just a conceptual device. They allow us to defer having to
decide whether to distinguish between byte and packet congestion when
the network resource codes the signal or when the transport decodes
it.
However, although this idealised mechanism isn't intended for
implementation, we do want to emphasise that we may need to find a
way to implement it, because it could become necessary to somehow
distinguish between bit and packet congestion [RFC3714]. Currently,
packet-congestion is not the common case, but there is no guarantee
that it will not become common with future technology trends.
The idealised wire protocol is given below. It accounts for packet
sizes at the transport layer, not in the network, and then only in
the case of bit-congestible resources. This avoids the perverse
incentive to send smaller packets and the DoS vulnerability that
would otherwise result if the network were to bias towards them (see
the motivating argument about avoiding perverse incentives in
Section 2.2):
1. A packet-congestible resource trying to code congestion level p_p
into a packet stream should mark the idealised `packet
congestion' field in each packet with probability p_p
irrespective of the packet's size. The transport should then
take a packet with the packet congestion field marked to mean
just one mark, irrespective of the packet size.
2. A bit-congestible resource trying to code time-varying byte-
congestion level p_b into a packet stream should mark the `byte
congestion' field in each packet with probability p_b, again
irrespective of the packet's size. Unlike before, the transport
should take a packet with the byte congestion field marked to
count as a mark on each byte in the packet.
The worked examples in Appendix B.2 show that transports can extract
sufficient and correct congestion notification from these protocols
for cases when two flows with different packet sizes have matching
bit rates or matching packet rates. Examples are also given that mix
these two flows into one to show that a flow with mixed packet sizes
would still be able to extract sufficient and correct information.
Sufficient and correct congestion information means that there is
sufficient information for the two different types of transport
requirements:
Ratio-based: Established transport congestion controls like TCP's
[RFC5681] aim to achieve equal segment rates per RTT through the
same bottleneck--TCP friendliness [RFC3448]. They work with the
ratio of dropped to delivered segments (or marked to unmarked
segments in the case of ECN). The example scenarios show that
these ratio-based transports are effectively the same whether
counting in bytes or packets, because the units cancel out.
(Incidentally, this is why TCP's bit rate is still proportional to
packet size even when byte-counting is used, as recommended for
TCP in [RFC5681], mainly for orthogonal security reasons.)
Absolute-target-based: Other congestion controls proposed in the
research community aim to limit the volume of congestion caused to
a constant weight parameter. [MulTCP][WindowPropFair] are
examples of weighted proportionally fair transports designed for
cost-fair environments [Rate_fair_Dis]. In this case, the
transport requires a count (not a ratio) of dropped/marked bytes
in the bit-congestible case and of dropped/marked packets in the
packet congestible case.
B.2. Example Scenarios
B.2.1. Notation
To prove our idealised wire protocol (Appendix B.1) is correct, we
will compare two flows with different packet sizes, s_1 and s_2 [bit/
pkt], to make sure their transports each see the correct congestion
notification. Initially, within each flow we will take all packets notification. Initially, within each flow we will take all packets
as having equal sizes, but later we will generalise to flows within as having equal sizes, but later we will generalise to flows within
which packet sizes vary. A flow's bit rate, x [bit/s], is related to which packet sizes vary. A flow's bit rate, x [bit/s], is related to
its packet rate, u [pkt/s], by its packet rate, u [pkt/s], by
x(t) = s.u(t). x(t) = s.u(t).
We will consider a 2x2 matrix of four scenarios: We will consider a 2x2 matrix of four scenarios:
+-----------------------------+------------------+------------------+ +-----------------------------+------------------+------------------+
| resource type and | A) Equal bit | B) Equal pkt | | resource type and | A) Equal bit | B) Equal pkt |
| congestion level | rates | rates | | congestion level | rates | rates |
+-----------------------------+------------------+------------------+ +-----------------------------+------------------+------------------+
| i) bit-congestible, p_b | (Ai) | (Bi) | | i) bit-congestible, p_b | (Ai) | (Bi) |
| ii) pkt-congestible, p_p | (Aii) | (Bii) | | ii) pkt-congestible, p_p | (Aii) | (Bii) |
+-----------------------------+------------------+------------------+ +-----------------------------+------------------+------------------+
Table 3 Table 3
A.2. Bit-congestible resource, equal bit rates (Ai) B.2.2. Bit-congestible resource, equal bit rates (Ai)
Starting with the bit-congestible scenario, for two flows to maintain Starting with the bit-congestible scenario, for two flows to maintain
equal bit rates (Ai) the ratio of the packet rates must be the equal bit rates (Ai) the ratio of the packet rates must be the
inverse of the ratio of packet sizes: u_2/u_1 = s_1/s_2. So, for inverse of the ratio of packet sizes: u_2/u_1 = s_1/s_2. So, for
instance, a flow of 60B packets would have to send 25x more packets instance, a flow of 60B packets would have to send 25x more packets
to achieve the same bit rate as a flow of 1500B packets. If a to achieve the same bit rate as a flow of 1500B packets. If a
congested resource marks proportion p_b of packets irrespective of congested resource marks proportion p_b of packets irrespective of
size, the ratio of marked packets received by each transport will size, the ratio of marked packets received by each transport will
still be the same as the ratio of their packet rates, p_b.u_2/p_b.u_1 still be the same as the ratio of their packet rates, p_b.u_2/p_b.u_1
= s_1/s_2. So of the 25x more 60B packets sent, 25x more will be = s_1/s_2. So of the 25x more 60B packets sent, 25x more will be
marked than in the 1500B packet flow, but 25x more won't be marked marked than in the 1500B packet flow, but 25x more won't be marked
too. too.
In this scenario, the resource is bit-congestible, so it always uses In this scenario, the resource is bit-congestible, so it always uses
our idealised bit-congestion field when it marks packets. Therefore our idealised bit-congestion field when it marks packets. Therefore
the transport should count marked bytes not packets. But it doesn't the transport should count marked bytes not packets. But it doesn't
actually matter for ratio-based transports like TCP (Section 5). The actually matter for ratio-based transports like TCP (Appendix B.1).
ratio of marked to unmarked bytes seen by each flow will be p_b, as The ratio of marked to unmarked bytes seen by each flow will be p_b,
will the ratio of marked to unmarked packets. Because they are as will the ratio of marked to unmarked packets. Because they are
ratios, the units cancel out. ratios, the units cancel out.
If a flow sent an inconsistent mixture of packet sizes, we have said If a flow sent an inconsistent mixture of packet sizes, we have said
it should count the ratio of marked and unmarked bytes not packets in it should count the ratio of marked and unmarked bytes not packets in
order to correctly decode the level of congestion. But actually, if order to correctly decode the level of congestion. But actually, if
all it is trying to do is decode p_b, it still doesn't matter. For all it is trying to do is decode p_b, it still doesn't matter. For
instance, imagine the two equal bit rate flows were actually one flow instance, imagine the two equal bit rate flows were actually one flow
at twice the bit rate sending a mixture of one 1500B packet for every at twice the bit rate sending a mixture of one 1500B packet for every
thirty 60B packets. 25x more small packets will be marked and 25x thirty 60B packets. 25x more small packets will be marked and 25x
more will be unmarked. The transport can still calculate p_b whether more will be unmarked. The transport can still calculate p_b whether
it uses bytes or packets for the ratio. In general, for any it uses bytes or packets for the ratio. In general, for any
algorithm which works on a ratio of marks to non-marks, either bytes algorithm which works on a ratio of marks to non-marks, either bytes
or packets can be counted interchangeably, because the choice cancels or packets can be counted interchangeably, because the choice cancels
out in the ratio calculation. out in the ratio calculation.
However, where an absolute target rather than relative volume of However, where an absolute target rather than relative volume of
congestion caused is important (Section 5), as it is for congestion congestion caused is important (Appendix B.1), as it is for
accountability [Rate_fair_Dis], the transport must count marked bytes congestion accountability [Rate_fair_Dis], the transport must count
not packets, in this bit-congestible case. Aside from the goal of marked bytes not packets, in this bit-congestible case. Aside from
congestion accountability, this is how the bit rate of a transport the goal of congestion accountability, this is how the bit rate of a
can be made independent of packet size; by ensuring the rate of transport can be made independent of packet size; by ensuring the
congestion caused is kept to a constant weight [WindowPropFair], rate of congestion caused is kept to a constant weight
rather than merely responding to the ratio of marked and unmarked [WindowPropFair], rather than merely responding to the ratio of
bytes. marked and unmarked bytes.
Note the unit of byte-congestion-volume is the byte. Note the unit of byte-congestion-volume is the byte.
A.3. Bit-congestible resource, equal packet rates (Bi) B.2.3. Bit-congestible resource, equal packet rates (Bi)
If two flows send different packet sizes but at the same packet rate, If two flows send different packet sizes but at the same packet rate,
their bit rates will be in the same ratio as their packet sizes, x_2/ their bit rates will be in the same ratio as their packet sizes, x_2/
x_1 = s_2/s_1. For instance, a flow sending 1500B packets at the x_1 = s_2/s_1. For instance, a flow sending 1500B packets at the
same packet rate as another sending 60B packets will be sending at same packet rate as another sending 60B packets will be sending at
25x greater bit rate. In this case, if a congested resource marks 25x greater bit rate. In this case, if a congested resource marks
proportion p_b of packets irrespective of size, the ratio of packets proportion p_b of packets irrespective of size, the ratio of packets
received with the byte-congestion field marked by each transport will received with the byte-congestion field marked by each transport will
be the same, p_b.u_2/p_b.u_1 = 1. be the same, p_b.u_2/p_b.u_1 = 1.
skipping to change at page 34, line 31 skipping to change at page 34, line 29
If the two flows are mixed into one, of bit rate x1+x2, with equal If the two flows are mixed into one, of bit rate x1+x2, with equal
packet rates of each size packet, the ratio p_b will still be packet rates of each size packet, the ratio p_b will still be
measurable by counting the ratio of marked to unmarked bytes (or measurable by counting the ratio of marked to unmarked bytes (or
packets because the ratio cancels out the units). However, if the packets because the ratio cancels out the units). However, if the
absolute volume of congestion is required, the transport must count absolute volume of congestion is required, the transport must count
the sum of congestion marked bytes, which indeed gives a correct the sum of congestion marked bytes, which indeed gives a correct
measure of the rate of byte-congestion p_b(x_1 + x_2) caused by the measure of the rate of byte-congestion p_b(x_1 + x_2) caused by the
combined bit rate. combined bit rate.
A.4. Pkt-congestible resource, equal bit rates (Aii) B.2.4. Pkt-congestible resource, equal bit rates (Aii)
Moving to the case of packet-congestible resources, we now take two Moving to the case of packet-congestible resources, we now take two
flows that send different packet sizes at the same bit rate, but this flows that send different packet sizes at the same bit rate, but this
time the pkt-congestion field is marked by the resource with time the pkt-congestion field is marked by the resource with
probability p_p. As in scenario Ai with the same bit rates but a probability p_p. As in scenario Ai with the same bit rates but a
bit-congestible resource, the flow with smaller packets will have a bit-congestible resource, the flow with smaller packets will have a
higher packet rate, so more packets will be both marked and unmarked, higher packet rate, so more packets will be both marked and unmarked,
but in the same proportion. but in the same proportion.
This time, the transport should only count marks without taking into This time, the transport should only count marks without taking into
skipping to change at page 35, line 13 skipping to change at page 35, line 10
flow of our example, as required. flow of our example, as required.
But if the transport is interested in the absolute number of packet But if the transport is interested in the absolute number of packet
congestion, it should just count how many marked packets arrive. For congestion, it should just count how many marked packets arrive. For
instance, a flow sending 60B packets will see 25x more marked packets instance, a flow sending 60B packets will see 25x more marked packets
than one sending 1500B packets at the same bit rate, because it is than one sending 1500B packets at the same bit rate, because it is
sending more packets through a packet-congestible resource. sending more packets through a packet-congestible resource.
Note the unit of packet congestion is a packet. Note the unit of packet congestion is a packet.
A.5. Pkt-congestible resource, equal packet rates (Bii) B.2.5. Pkt-congestible resource, equal packet rates (Bii)
Finally, if two flows with the same packet rate, pass through a Finally, if two flows with the same packet rate, pass through a
packet-congestible resource, they will both suffer the same packet-congestible resource, they will both suffer the same
proportion of marking, p_p, irrespective of their packet sizes. On proportion of marking, p_p, irrespective of their packet sizes. On
detecting that the pkt-congestion field is marked, the transport detecting that the pkt-congestion field is marked, the transport
should count packets, and it will be able to extract the ratio p_p of should count packets, and it will be able to extract the ratio p_p of
marked to unmarked packets from both flows, irrespective of packet marked to unmarked packets from both flows, irrespective of packet
sizes. sizes.
Even if the transport is monitoring the absolute amount of packets Even if the transport is monitoring the absolute amount of packets
congestion over a period, still it will see the same amount of packet congestion over a period, still it will see the same amount of packet
congestion from either flow. congestion from either flow.
And if the two equal packet rates of different size packets are mixed And if the two equal packet rates of different size packets are mixed
together in one flow, the packet rate will double, so the absolute together in one flow, the packet rate will double, so the absolute
volume of packet-congestion will accumulate at twice the rate of volume of packet-congestion will accumulate at twice the rate of
either flow, 2p_p.u_1 = p_p(u_1+u_2). either flow, 2p_p.u_1 = p_p(u_1+u_2).
Appendix B. Congestion Notification Definition: Further Justification
In Section 3 on the definition of congestion notification, load not
capacity was used as the denominator. This also has a subtle
significance in the related debate over the design of new transport
protocols--typical new protocol designs (e.g. in XCP [xcp-spec] &
Quickstart [RFC4782]) expect the sending transport to communicate its
desired flow rate to the network and network elements to
progressively subtract from this so that the achievable flow rate
emerges at the receiving transport.
Congestion notification with total load in the denominator can serve
a similar purpose (though in retrospect not in advance like XCP &
QuickStart). Congestion notification is a dimensionless fraction but
each source can extract necessary rate information from it because it
already knows what its own rate is. Even though congestion
notification doesn't communicate a rate explicitly, from each
source's point of view congestion notification represents the
fraction of the rate it was sending a round trip ago that couldn't
(or wouldn't) be served by available resources. After they were
sent, all these fractions of each source's offered load added up to
the aggregate fraction of offered load seen by the congested
resource. So, the source can also know the total excess rate by
multiplying total load by congestion level. Therefore congestion
notification, as one scale-free dimensionless fraction, implicitly
communicates the instantaneous excess flow rate, albeit a RTT ago.
Appendix C. Byte-mode Drop Complicates Policing Congestion Response Appendix C. Byte-mode Drop Complicates Policing Congestion Response
This appendix explains why the ability of networks to police the This appendix explains why the ability of networks to police the
response of _any_ transport to congestion depends on bit-congestible response of _any_ transport to congestion depends on bit-congestible
network resources only doing packet-mode not byte-mode drop. network resources only doing packet-mode not byte-mode drop.
To be able to police a transport's response to congestion when To be able to police a transport's response to congestion when
fairness can only be judged over time and over all an individual's fairness can only be judged over time and over all an individual's
flows, the policer has to have an integrated view of all the flows, the policer has to have an integrated view of all the
congestion an individual (not just one flow) has caused due to all congestion an individual (not just one flow) has caused due to all
traffic entering the Internet from that individual. This is termed traffic entering the Internet from that individual. This is termed
congestion accountability. congestion accountability.
But a byte-mode drop algorithm has to depend on the local MTU of the But a byte-mode drop algorithm has to depend on the local MTU of the
line - an algorithm needs to use some concept of a 'normal' packet line - an algorithm needs to use some concept of a 'normal' packet
size. Therefore, one dropped or marked packet is not necessarily size. Therefore, one dropped or marked packet is not necessarily
equivalent to another unless you know the MTU at the queue that where equivalent to another unless you know the MTU at the queue where it
it was dropped/marked. To have an integrated view of a user, we was dropped/marked. To have an integrated view of a user, we believe
believe congestion policing has to be located at an individual's congestion policing has to be located at an individual's attachment
attachment point to the Internet [I-D.briscoe-tsvwg-re-ecn-tcp]. But point to the Internet [I-D.briscoe-tsvwg-re-ecn-tcp]. But from there
from there it cannot know the MTU of each remote queue that caused it cannot know the MTU of each remote queue that caused each drop/
each drop/mark. Therefore it cannot take an integrated approach to mark. Therefore it cannot take an integrated approach to policing
policing all the responses to congestion of all the transports of one all the responses to congestion of all the transports of one
individual. Therefore it cannot police anything. individual. Therefore it cannot police anything.
The security/incentive argument _for_ packet-mode drop is similar. The security/incentive argument _for_ packet-mode drop is similar.
Firstly, confining RED to packet-mode drop would not preclude Firstly, confining RED to packet-mode drop would not preclude
bottleneck policing approaches such as [pBox] as it seems likely they bottleneck policing approaches such as [pBox] as it seems likely they
could work just as well by monitoring the volume of dropped bytes could work just as well by monitoring the volume of dropped bytes
rather than packets. Secondly packet-mode dropping/marking naturally rather than packets. Secondly packet-mode dropping/marking naturally
allows the congestion notification of packets to be globally allows the congestion notification of packets to be globally
meaningful without relying on MTU information held elsewhere. meaningful without relying on MTU information held elsewhere.
skipping to change at page 37, line 11 skipping to change at page 36, line 27
packets or across different size flows [Rate_fair_Dis]. Therefore packets or across different size flows [Rate_fair_Dis]. Therefore
policing would work naturally with just simple packet-mode drop in policing would work naturally with just simple packet-mode drop in
RED. RED.
In summary, making drop probability depend on the size of the packets In summary, making drop probability depend on the size of the packets
that bits happen to be divided into simply encourages the bits to be that bits happen to be divided into simply encourages the bits to be
divided into smaller packets. Byte-mode drop would therefore divided into smaller packets. Byte-mode drop would therefore
irreversibly complicate any attempt to fix the Internet's incentive irreversibly complicate any attempt to fix the Internet's incentive
structures. structures.
Author's Address Appendix D. Changes from Previous Versions
To be removed by the RFC Editor on publication.
Full incremental diffs between each version are available at
<http://www.cs.ucl.ac.uk/staff/B.Briscoe/pubs.html#byte-pkt-congest>
or
<http://tools.ietf.org/wg/tsvwg/draft-ietf-tsvwg-byte-pkt-congest/>
(courtesy of the rfcdiff tool):
From -01 to -02 (this version):
* Restructured the whole document for (hopefully) easier reading
and clarity. The concrete recommendation, in RFC2119 language,
is now in Section 5.
From -00 to -01:
* Minor clarifications throughout and updated references
From briscoe-byte-pkt-mark-02 to ietf-byte-pkt-congest-00:
* Added note on relationship to existing RFCs
* Posed the question of whether packet-congestion could become
common and deferred it to the IRTF ICCRG. Added ref to the
dual-resource queue (DRQ) proposal.
* Changed PCN references from the PCN charter & architecture to
the PCN marking behaviour draft most likely to imminently
become the standards track WG item.
From -01 to -02:
* Abstract reorganised to align with clearer separation of issue
in the memo.
* Introduction reorganised with motivating arguments removed to
new Section 2.
* Clarified avoiding lock-out of large packets is not the main or
only motivation for RED.
* Mentioned choice of drop or marking explicitly throughout,
rather than trying to coin a word to mean either.
* Generalised the discussion throughout to any packet forwarding
function on any network equipment, not just routers.
* Clarified the last point about why this is a good time to sort
out this issue: because it will be hard / impossible to design
new transports unless we decide whether the network or the
transport is allowing for packet size.
* Added statement explaining the horizon of the memo is long
term, but with short term expediency in mind.
* Added material on scaling congestion control with packet size
(Section 2.1).
* Separated out issue of normalising TCP's bit rate from issue of
preference to control packets (Section 2.3).
* Divided up Congestion Measurement section for clarity,
including new material on fixed size packet buffers and buffer
carving (Section 3.1.1 & Section 3.2.1) and on congestion
measurement in wireless link technologies without queues
(Section 3.1.2).
* Added section on 'Making Transports Robust against Control
Packet Losses' (Section 3.2.3) with existing & new material
included.
* Added tabulated results of vendor survey on byte-mode drop
variant of RED (Table 2).
From -00 to -01:
* Clarified applicability to drop as well as ECN.
* Highlighted DoS vulnerability.
* Emphasised that drop-tail suffers from similar problems to
byte-mode drop, so only byte-mode drop should be turned off,
not RED itself.
* Clarified the original apparent motivations for recommending
byte-mode drop included protecting SYNs and pure ACKs more than
equalising the bit rates of TCPs with different segment sizes.
Removed some conjectured motivations.
* Added support for updates to TCP in progress (ackcc & ecn-syn-
ack).
* Updated survey results with newly arrived data.
* Pulled all recommendations together into the conclusions.
* Moved some detailed points into two additional appendices and a
note.
* Considerable clarifications throughout.
* Updated references
Authors' Addresses
Bob Briscoe Bob Briscoe
BT BT
B54/77, Adastral Park B54/77, Adastral Park
Martlesham Heath Martlesham Heath
Ipswich IP5 3RE Ipswich IP5 3RE
UK UK
Phone: +44 1473 645196 Phone: +44 1473 645196
Email: bob.briscoe@bt.com EMail: bob.briscoe@bt.com
URI: http://bobbriscoe.net/ URI: http://bobbriscoe.net/
Jukka Manner
Aalto University
Department of Communications and Networking (Comnet)
P.O. Box 13000
FIN-00076 Aalto
Finland
Phone: +358 9 470 22481
EMail: jukka.manner@tkk.fi
URI: http://www.netlab.tkk.fi/~jmanner/
 End of changes. 150 change blocks. 
846 lines changed or deleted 919 lines changed or added

This html diff was produced by rfcdiff 1.38. The latest version is available from http://tools.ietf.org/tools/rfcdiff/