draft-ietf-tsvwg-aqm-dualq-coupled-00.txt   draft-ietf-tsvwg-aqm-dualq-coupled-01.txt 
Transport Area working group (tsvwg) K. De Schepper Transport Area working group (tsvwg) K. De Schepper
Internet-Draft Nokia Bell Labs Internet-Draft Nokia Bell Labs
Intended status: Experimental B. Briscoe, Ed. Intended status: Experimental B. Briscoe, Ed.
Expires: November 11, 2017 O. Bondarenko Expires: January 4, 2018 O. Bondarenko
Simula Research Lab Simula Research Lab
I. Tsang I. Tsang
Nokia Bell Labs Nokia Bell Labs
May 10, 2017 July 3, 2017
DualQ Coupled AQM for Low Latency, Low Loss and Scalable Throughput DualQ Coupled AQM for Low Latency, Low Loss and Scalable Throughput
draft-ietf-tsvwg-aqm-dualq-coupled-00 draft-ietf-tsvwg-aqm-dualq-coupled-01
Abstract Abstract
Data Centre TCP (DCTCP) was designed to provide predictably low Data Centre TCP (DCTCP) was designed to provide predictably low
queuing latency, near-zero loss, and throughput scalability using queuing latency, near-zero loss, and throughput scalability using
explicit congestion notification (ECN) and an extremely simple explicit congestion notification (ECN) and an extremely simple
marking behaviour on switches. However, DCTCP does not co-exist with marking behaviour on switches. However, DCTCP does not co-exist with
existing TCP traffic---throughput starves. So, until now, DCTCP existing TCP traffic---DCTCP is so aggressive that existing TCP
could only be deployed where a clean-slate environment could be algorithms approach starvation. So, until now, DCTCP could only be
arranged, such as in private data centres. This specification deployed where a clean-slate environment could be arranged, such as
defines `DualQ Coupled Active Queue Management (AQM)' to allow in private data centres. This specification defines `DualQ Coupled
scalable congestion controls like DCTCP to safely co-exist with Active Queue Management (AQM)' to allow scalable congestion controls
classic Internet traffic. The Coupled AQM ensures that a flow runs like DCTCP to safely co-exist with classic Internet traffic. The
at about the same rate whether it uses DCTCP or TCP Reno/Cubic, but Coupled AQM ensures that a flow runs at about the same rate whether
without inspecting transport layer flow identifiers. When tested in it uses DCTCP or TCP Reno/Cubic, but without inspecting transport
a residential broadband setting, DCTCP achieved sub-millisecond layer flow identifiers. When tested in a residential broadband
average queuing delay and zero congestion loss under a wide range of setting, DCTCP achieved sub-millisecond average queuing delay and
mixes of DCTCP and `Classic' broadband Internet traffic, without zero congestion loss under a wide range of mixes of DCTCP and
compromising the performance of the Classic traffic. The solution `Classic' broadband Internet traffic, without compromising the
also reduces network complexity and eliminates network configuration. performance of the Classic traffic. The solution also reduces
network complexity and eliminates network configuration.
Status of This Memo Status of This Memo
This Internet-Draft is submitted in full conformance with the 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). Note that other groups may also distribute Task Force (IETF). Note that other groups may also distribute
working documents as Internet-Drafts. The list of current Internet- working documents as Internet-Drafts. The list of current Internet-
Drafts is at http://datatracker.ietf.org/drafts/current/. 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."
This Internet-Draft will expire on November 11, 2017. This Internet-Draft will expire on January 4, 2018.
Copyright Notice Copyright Notice
Copyright (c) 2017 IETF Trust and the persons identified as the Copyright (c) 2017 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 Provisions Relating to IETF Documents
(http://trustee.ietf.org/license-info) in effect on the date of (http://trustee.ietf.org/license-info) in effect on the date of
publication of this document. Please review these documents publication of this document. Please review these documents
carefully, as they describe your rights and restrictions with respect carefully, as they describe your rights and restrictions with respect
to this document. Code Components extracted from this document must to this document. Code Components extracted from this document must
include Simplified BSD License text as described in Section 4.e of include Simplified BSD License text as described in Section 4.e of
the Trust Legal Provisions and are provided without warranty as the Trust Legal Provisions and are provided without warranty as
described in the Simplified BSD License. described in the Simplified BSD License.
Table of Contents Table of Contents
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2
1.1. Problem and Scope . . . . . . . . . . . . . . . . . . . . 2 1.1. Problem and Scope . . . . . . . . . . . . . . . . . . . . 3
1.2. Terminology . . . . . . . . . . . . . . . . . . . . . . . 5 1.2. Terminology . . . . . . . . . . . . . . . . . . . . . . . 5
1.3. Features . . . . . . . . . . . . . . . . . . . . . . . . 5 1.3. Features . . . . . . . . . . . . . . . . . . . . . . . . 5
2. DualQ Coupled AQM Algorithm . . . . . . . . . . . . . . . . . 6 2. DualQ Coupled AQM Algorithm . . . . . . . . . . . . . . . . . 7
2.1. Coupled AQM . . . . . . . . . . . . . . . . . . . . . . . 7 2.1. Coupled AQM . . . . . . . . . . . . . . . . . . . . . . . 7
2.2. Dual Queue . . . . . . . . . . . . . . . . . . . . . . . 8 2.2. Dual Queue . . . . . . . . . . . . . . . . . . . . . . . 8
2.3. Traffic Classification . . . . . . . . . . . . . . . . . 8 2.3. Traffic Classification . . . . . . . . . . . . . . . . . 8
2.4. Normative Requirements . . . . . . . . . . . . . . . . . 8 2.4. Normative Requirements . . . . . . . . . . . . . . . . . 8
3. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 9 3. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 9
4. Security Considerations . . . . . . . . . . . . . . . . . . . 9 4. Security Considerations . . . . . . . . . . . . . . . . . . . 10
4.1. Overload Handling . . . . . . . . . . . . . . . . . . . . 10 4.1. Overload Handling . . . . . . . . . . . . . . . . . . . . 10
5. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 11 5. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 11
6. References . . . . . . . . . . . . . . . . . . . . . . . . . 11 6. References . . . . . . . . . . . . . . . . . . . . . . . . . 11
6.1. Normative References . . . . . . . . . . . . . . . . . . 11 6.1. Normative References . . . . . . . . . . . . . . . . . . 11
6.2. Informative References . . . . . . . . . . . . . . . . . 11 6.2. Informative References . . . . . . . . . . . . . . . . . 11
Appendix A. Example DualQ Coupled PI2 Algorithm . . . . . . . . 14 Appendix A. Example DualQ Coupled PI2 Algorithm . . . . . . . . 14
Appendix B. Example DualQ Coupled Curvy RED Algorithm . . . . . 17 A.1. Pass #1: Core Concepts . . . . . . . . . . . . . . . . . 15
Appendix C. Guidance on Controlling Throughput Equivalence . . . 23 A.2. Pass #2: Overload Details . . . . . . . . . . . . . . . . 18
Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 24 Appendix B. Example DualQ Coupled Curvy RED Algorithm . . . . . 21
Appendix C. Guidance on Controlling Throughput Equivalence . . . 26
Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 28
1. Introduction 1. Introduction
1.1. Problem and Scope 1.1. Problem and Scope
Latency is becoming the critical performance factor for many (most?) Latency is becoming the critical performance factor for many (most?)
applications on the public Internet, e.g. interactive Web, Web applications on the public Internet, e.g. interactive Web, Web
services, voice, conversational video, interactive video, interactive services, voice, conversational video, interactive video, interactive
remote presence, instant messaging, online gaming, remote desktop, remote presence, instant messaging, online gaming, remote desktop,
cloud-based applications, and video-assisted remote control of cloud-based applications, and video-assisted remote control of
machinery and industrial processes. In the developed world, further machinery and industrial processes. In the developed world, further
increases in access network bit-rate offer diminishing returns, increases in access network bit-rate offer diminishing returns,
whereas latency is still a multi-faceted problem. In the last decade whereas latency is still a multi-faceted problem. In the last decade
skipping to change at page 3, line 38 skipping to change at page 3, line 44
some traffic at the expense of others, AQM controls latency for _all_ some traffic at the expense of others, AQM controls latency for _all_
traffic in a class. In general, AQMs introduce an increasing level traffic in a class. In general, AQMs introduce an increasing level
of discard from the buffer the longer the queue persists above a of discard from the buffer the longer the queue persists above a
shallow threshold. This gives sufficient signals to capacity-seeking shallow threshold. This gives sufficient signals to capacity-seeking
(aka. greedy) flows to keep the buffer empty for its intended (aka. greedy) flows to keep the buffer empty for its intended
purpose: absorbing bursts. However, RED [RFC2309] and other purpose: absorbing bursts. However, RED [RFC2309] and other
algorithms from the 1990s were sensitive to their configuration and algorithms from the 1990s were sensitive to their configuration and
hard to set correctly. So, AQM was not widely deployed. hard to set correctly. So, AQM was not widely deployed.
More recent state-of-the-art AQMs, e.g. More recent state-of-the-art AQMs, e.g.
fq_CoDel [I-D.ietf-aqm-fq-codel], PIE [I-D.ietf-aqm-pie], Adaptive fq_CoDel [I-D.ietf-aqm-fq-codel], PIE [RFC8033], Adaptive
RED [ARED01], are easier to configure, because they define the RED [ARED01], are easier to configure, because they define the
queuing threshold in time not bytes, so it is invariant for different queuing threshold in time not bytes, so it is invariant for different
link rates. However, no matter how good the AQM, the sawtoothing link rates. However, no matter how good the AQM, the sawtoothing
rate of TCP will either cause queuing delay to vary or cause the link rate of TCP will either cause queuing delay to vary or cause the link
to be under-utilized. Even with a perfectly tuned AQM, the to be under-utilized. Even with a perfectly tuned AQM, the
additional queuing delay will be of the same order as the underlying additional queuing delay will be of the same order as the underlying
speed-of-light delay across the network. Flow-queuing can isolate speed-of-light delay across the network. Flow-queuing can isolate
one flow from another, but it cannot isolate a TCP flow from the one flow from another, but it cannot isolate a TCP flow from the
delay variations it inflicts on itself, and it has other problems - delay variations it inflicts on itself, and it has other problems -
it overrides the flow rate decisions of variable rate video it overrides the flow rate decisions of variable rate video
skipping to change at page 4, line 46 skipping to change at page 5, line 5
the problem of coexistence between scalable and classic flows, the problem of coexistence between scalable and classic flows,
without having to inspect flow identifiers. The AQM is not like without having to inspect flow identifiers. The AQM is not like
flow-queuing approaches [I-D.ietf-aqm-fq-codel] that classify packets flow-queuing approaches [I-D.ietf-aqm-fq-codel] that classify packets
by flow identifier into numerous separate queues in order to isolate by flow identifier into numerous separate queues in order to isolate
sparse flows from the higher latency in the queues assigned to sparse flows from the higher latency in the queues assigned to
heavier flow. In contrast, the AQM exploits the behaviour of heavier flow. In contrast, the AQM exploits the behaviour of
scalable congestion controls like DCTCP so that every packet in every scalable congestion controls like DCTCP so that every packet in every
flow sharing the queue for DCTCP-like traffic can be served with very flow sharing the queue for DCTCP-like traffic can be served with very
low latency. low latency.
This AQM extension can be combined with any single qeueu AQM that This AQM extension can be combined with any single queue AQM that
generates a statistical or deterministic mark/drop probability driven generates a statistical or deterministic mark/drop probability driven
by the queue dynamics. In many cases it simplifies the basic control by the queue dynamics. In many cases it simplifies the basic control
algorithm, and requires little extra processing. Therefore it is algorithm, and requires little extra processing. Therefore it is
believed the Coupled AQM would be applicable and easy to deploy in believed the Coupled AQM would be applicable and easy to deploy in
all types of buffers; buffers in cost-reduced mass-market residential all types of buffers; buffers in cost-reduced mass-market residential
equipment; buffers in end-system stacks; buffers in carrier-scale equipment; buffers in end-system stacks; buffers in carrier-scale
equipment including remote access servers, routers, firewalls and equipment including remote access servers, routers, firewalls and
Ethernet switches; buffers in network interface cards, buffers in Ethernet switches; buffers in network interface cards, buffers in
virtualized network appliances, hypervisors, and so on. virtualized network appliances, hypervisors, and so on.
skipping to change at page 7, line 46 skipping to change at page 8, line 7
There is a really simple way to implement the square of a probability There is a really simple way to implement the square of a probability
- by testing the queue against two random numbers not one. This is - by testing the queue against two random numbers not one. This is
the approach adopted in Appendix A and Appendix B. the approach adopted in Appendix A and Appendix B.
Stating this as a formula, the relation between Classic drop Stating this as a formula, the relation between Classic drop
probability, p_C, and L4S marking probability, p_L needs to take the probability, p_C, and L4S marking probability, p_L needs to take the
form: form:
p_C = ( p_L / k )^2 (1) p_C = ( p_L / k )^2 (1)
where k is the constant of proportionality. Optionally, k can be where k is the constant of proportionality.
expressed as a power of 2, so k=2^k', where k' is another constant.
Then implementations can avoid costly division by shifting p_L by k'
bits to the right.
2.2. Dual Queue 2.2. Dual Queue
Classic traffic builds a large queue, so a separate queue is provided Classic traffic builds a large queue, so a separate queue is provided
for L4S traffic, and it is scheduled with strict priority. for L4S traffic, and it is scheduled with strict priority.
Nonetheless, coupled marking ensures that giving priority to L4S Nonetheless, coupled marking ensures that giving priority to L4S
traffic still leaves the right amount of spare scheduling time for traffic still leaves the right amount of spare scheduling time for
Classic flows to each get equivalent throughput to DCTCP flows (all Classic flows to each get equivalent throughput to DCTCP flows (all
other factors such as RTT being equal). The algorithm achieves this other factors such as RTT being equal). The algorithm achieves this
without having to inspect flow identifiers. without having to inspect flow identifiers.
skipping to change at page 9, line 31 skipping to change at page 9, line 34
Typically, access network operators isolate customers from each other Typically, access network operators isolate customers from each other
with some form of layer-2 multiplexing (TDM in DOCSIS, CDMA in 3G) or with some form of layer-2 multiplexing (TDM in DOCSIS, CDMA in 3G) or
L3 scheduling (WRR in broadband), rather than relying on TCP to share L3 scheduling (WRR in broadband), rather than relying on TCP to share
capacity between customers [RFC0970]. In such cases, the choice of k capacity between customers [RFC0970]. In such cases, the choice of k
will solely affect relative flow rates within each customer's access will solely affect relative flow rates within each customer's access
capacity, not between customers. Also, k will not affect relative capacity, not between customers. Also, k will not affect relative
flow rates at any times when all flows are Classic or all L4S, and it flow rates at any times when all flows are Classic or all L4S, and it
will not affect small flows. will not affect small flows.
Example DualQ Coupled AQM algorithms called PI2 and Curvy RED are Example DualQ Coupled AQM algorithms called DualPI2 and Curvy RED are
given in Appendix A and Appendix B. Either example AQM can be used given in Appendix A and Appendix B. Either example AQM can be used
to couple packet marking and dropping across a dual Q. Curvy RED to couple packet marking and dropping across a dual Q. Curvy RED
requires less operations per packet than RED and can be used if the requires less operations per packet than RED and can be used if the
range of RTTs is limited. PI2 is a simplification of PIE with stable range of RTTs is limited. DualPI2 is a simplification of PIE with
Proportional-Integral control for both Classic and L4S congestion stable Proportional-Integral control for both Classic and L4S
controls. Nonetheless, it would be possible to control the queues congestion controls. Nonetheless, it would be possible to control
with other alternative AQMs, as long as the above normative the queues with other alternative AQMs, as long as the above
requirements (those expressed in capitals) are observed, which are normative requirements (those expressed in capitals) are observed,
intended to be independent of the specific AQM. which are intended to be independent of the specific AQM.
{ToDo: Add management and monitoring requirements} {ToDo: Add management and monitoring requirements}
3. IANA Considerations 3. IANA Considerations
This specification contains no IANA considerations. This specification contains no IANA considerations.
4. Security Considerations 4. Security Considerations
4.1. Overload Handling 4.1. Overload Handling
Where the interests of users or flows might conflict, it could be Where the interests of users or flows might conflict, it could be
necessary to police traffic to isolate any harm to performance. This necessary to police traffic to isolate any harm to performance. This
is a policy issue that needs to be separable from a basic AQM, but an is a policy issue that needs to be separable from a basic AQM, but an
AQM does need to handle overload. A trade-off needs to be made AQM does need to handle overload. A trade-off needs to be made
between complexity and the risk of either class harming the other. between complexity and the risk of either class harming the other.
It is an operator policy to define what must happen if the service It is an operator policy to define what must happen if the service
time of the classic queue becomes too great. In the following time of the classic queue becomes too great. In the following
subsections three optional non-exclusive overload protections are subsections three optional non-exclusive overload protections are
skipping to change at page 12, line 29 skipping to change at page 12, line 34
(Under submission) (Under submission)
[I-D.ietf-aqm-fq-codel] [I-D.ietf-aqm-fq-codel]
Hoeiland-Joergensen, T., McKenney, P., Hoeiland-Joergensen, T., McKenney, P.,
dave.taht@gmail.com, d., Gettys, J., and E. Dumazet, "The dave.taht@gmail.com, d., Gettys, J., and E. Dumazet, "The
FlowQueue-CoDel Packet Scheduler and Active Queue FlowQueue-CoDel Packet Scheduler and Active Queue
Management Algorithm", draft-ietf-aqm-fq-codel-06 (work in Management Algorithm", draft-ietf-aqm-fq-codel-06 (work in
progress), March 2016. progress), March 2016.
[I-D.ietf-aqm-pie]
Pan, R., Natarajan, P., Baker, F., and G. White, "PIE: A
Lightweight Control Scheme To Address the Bufferbloat
Problem", draft-ietf-aqm-pie-10 (work in progress),
September 2016.
[I-D.ietf-tcpm-cubic] [I-D.ietf-tcpm-cubic]
Rhee, I., Xu, L., Ha, S., Zimmermann, A., Eggert, L., and Rhee, I., Xu, L., Ha, S., Zimmermann, A., Eggert, L., and
R. Scheffenegger, "CUBIC for Fast Long-Distance Networks", R. Scheffenegger, "CUBIC for Fast Long-Distance Networks",
draft-ietf-tcpm-cubic-04 (work in progress), February draft-ietf-tcpm-cubic-04 (work in progress), February
2017. 2017.
[I-D.ietf-tcpm-dctcp] [I-D.ietf-tcpm-dctcp]
Bensley, S., Thaler, D., Balasubramanian, P., Eggert, L., Bensley, S., Thaler, D., Balasubramanian, P., Eggert, L.,
and G. Judd, "Datacenter TCP (DCTCP): TCP Congestion and G. Judd, "Datacenter TCP (DCTCP): TCP Congestion
Control for Datacenters", draft-ietf-tcpm-dctcp-05 (work Control for Datacenters", draft-ietf-tcpm-dctcp-08 (work
in progress), March 2017. in progress), June 2017.
[I-D.ietf-tsvwg-ecn-experimentation] [I-D.ietf-tsvwg-ecn-experimentation]
Black, D., "Explicit Congestion Notification (ECN) Black, D., "Explicit Congestion Notification (ECN)
Experimentation", draft-ietf-tsvwg-ecn-experimentation-00 Experimentation", draft-ietf-tsvwg-ecn-experimentation-00
(work in progress), November 2016. (work in progress), November 2016.
[I-D.ietf-tsvwg-ecn-l4s-id] [I-D.ietf-tsvwg-ecn-l4s-id]
Schepper, K., Briscoe, B., and I. Tsang, "Identifying Schepper, K., Briscoe, B., and I. Tsang, "Identifying
Modified Explicit Congestion Notification (ECN) Semantics Modified Explicit Congestion Notification (ECN) Semantics
for Ultra-Low Queuing Delay", draft-ietf-tsvwg-ecn-l4s- for Ultra-Low Queuing Delay", draft-ietf-tsvwg-ecn-l4s-
skipping to change at page 14, line 24 skipping to change at page 14, line 24
<http://www.rfc-editor.org/info/rfc3246>. <http://www.rfc-editor.org/info/rfc3246>.
[RFC3649] Floyd, S., "HighSpeed TCP for Large Congestion Windows", [RFC3649] Floyd, S., "HighSpeed TCP for Large Congestion Windows",
RFC 3649, DOI 10.17487/RFC3649, December 2003, RFC 3649, DOI 10.17487/RFC3649, December 2003,
<http://www.rfc-editor.org/info/rfc3649>. <http://www.rfc-editor.org/info/rfc3649>.
[RFC5681] Allman, M., Paxson, V., and E. Blanton, "TCP Congestion [RFC5681] Allman, M., Paxson, V., and E. Blanton, "TCP Congestion
Control", RFC 5681, DOI 10.17487/RFC5681, September 2009, Control", RFC 5681, DOI 10.17487/RFC5681, September 2009,
<http://www.rfc-editor.org/info/rfc5681>. <http://www.rfc-editor.org/info/rfc5681>.
[RFC7567] Baker, F., Ed. and G. Fairhurst, Ed., "IETF
Recommendations Regarding Active Queue Management",
BCP 197, RFC 7567, DOI 10.17487/RFC7567, July 2015,
<http://www.rfc-editor.org/info/rfc7567>.
[RFC8033] Pan, R., Natarajan, P., Baker, F., and G. White,
"Proportional Integral Controller Enhanced (PIE): A
Lightweight Control Scheme to Address the Bufferbloat
Problem", RFC 8033, DOI 10.17487/RFC8033, February 2017,
<http://www.rfc-editor.org/info/rfc8033>.
Appendix A. Example DualQ Coupled PI2 Algorithm Appendix A. Example DualQ Coupled PI2 Algorithm
As a first concrete example, the pseudocode below gives the DualQ As a first concrete example, the pseudocode below gives the DualPI2
Coupled AQM algorithm based on the PI2 Classic AQM, we used and algorithm, which is a DualQ Coupled AQM algorithm based on the PI2
tested. For this example only the pseudo code is given. An open algorithm [PI216] for the Classic AQM. Pi2 is an improved variant of
source implementation for Linux is available at: the PIE AQM [RFC8033].
We will introduce the pseudocode in two passes. The first pass
explains the core concepts, deferring handling of overload to the
second pass. The first pass also uses regular arithmetic, whereas
some integer arithmetic more suitable for kernel operations appears
in the second pass. To aid comparison, line numbers are kept in step
between the two passes by using letter suffixes where the longer code
needs extra lines.
A full open source implementation for Linux is available at:
https://github.com/olgabo/dualpi2. https://github.com/olgabo/dualpi2.
1: dualpi2_enqueue(lq, cq, pkt) { % Test limit and classify lq or cq A.1. Pass #1: Core Concepts
2: stamp(pkt) % attach arrival time to packet
3: if ( lq.len() + cq.len() > limit )
4: drop(pkt) % drop packet if q is full
5: else {
6: if ( ecn(pkt) modulo 2 == 0 ) % ECN bits = not-ect or ect(0)
7: cq.enqueue(pkt)
8: else % ECN bits = ect(1) or ce
9: lq.enqueue(pkt)
10: }
11: }
Figure 1: Example Enqueue Pseudocode for DualQ Coupled PI2 AQM The pseudocode manipulates three main structures of variables: the
packet (pkt), the L4S queue (lq) and the Classic queue (cq). The
pseudocode consists of the following four functions:
1: dualpi2_dequeue(lq, cq) { % Couples L4S & Classic queues, lq & cq o initialization code (Figure 1) that sets parameter defaults (the
2: while ( lq.len() + cq.len() > 0 ) API for setting non-default values is omitted for brevity)
3: if ( lq.time() + tshift >= cq.time() ) {
4: lq.dequeue(pkt)
5: if ( (pkt.time() > T) or (p * k > rand()) ) % coupling here
6: mark(pkt)
7: return(pkt) % return the packet and stop here
8: } else {
9: cq.dequeue(pkt)
10: if ( p > max(rand(), rand()) ) % only square part of (p/k)^2
11: if ( ecn(pkt) == 0 ) % ECN field = not-ect
12: drop(pkt) % squared drop, redo loop
13: else {
14: mark(pkt) % squared mark
15: return(pkt) % return the packet and stop here
16: }
17: else
18: return(pkt) % return the packet and stop here
19: }
20: }
21: return(NULL) % no packet to dequeue
22: }
Figure 2: Example Dequeue Pseudocode for DualQ Coupled PI2 AQM o enqueue code (Figure 2)
1: dualpi2_update(lq, cq) { % Update p every Tupdate o dequeue code (Figure 3)
2: curq = cq.time() % use queuing time of first-in Classic packet
3: alpha_U = alpha * Tupdate % done once when parameters are set
4: beta_U = beta * Tupdate % done once when parameters are set
5: p = p + alpha_U * (curq - target) + beta_U * (curq - prevq)
6: prevq = curq
7: }
Figure 3: Example PI-Update Pseudocode for DualQ Coupled PI2 AQM o code to regularly update the base probability (p) used in the
dequeue code (Figure 4).
The base probability (p) is an internal variable from which the
marking and dropping probabilities for L4S and Classic traffic (p_L
and p_C) are derived. Specifically, p_L = p * k and p_C = p^2, from
Equation (1) (see lines 4a and 4b in Figure 4).
In our experiments so far (building on experiments with PIE) on
broadband access links ranging from 4 Mb/s to 200 Mb/s with base RTTs
from 5 ms to 100 ms, DualPI2 achieves good results with the default
parameters in Figure 1.
1: dualpi2_params_init(...) { % Set parameter defaults
2: target = 15 ms % PI AQM Classic queue delay target
3: tshift = 2 * target % Scheduler time bias
4: Tupdate = 16 ms % PI Classic queue sampling interval
5: alpha = 10 Hz^2 % PI integral gain
6: beta = 100 Hz^2 % PI proportional gain
7: alpha_U = alpha *Tupdate % PI integral gain per update interval
8: beta_U = beta * Tupdate % PI prop'nal gain per update interval
9: T_time = 1 ms % L4S marking threshold in time
10: T_len = 2 * MTU % Min L4S marking threshold in bytes
11: k = 2 % Coupling factor
12: limit = MAX_LINK_RATE * 250 ms % Dual buffer size
13: p_Cmax = 1/4 % Max Classic drop/mark prob
14: p_Lmax = min(k*sqrt(p_Cmax), 1) % Max L4S marking prob
15: }
Figure 1: Example Header Pseudocode for DualQ Coupled PI2 AQM
1: dualpi2_enqueue(lq, cq, pkt) { % Test limit and classify lq or cq
2: stamp(pkt) % attach arrival time to packet
3: if ( lq.len() + cq.len() > limit )
4: drop(pkt) % drop packet if buffer is full
5: else { % Packet classifier
6: if ( ecn(pkt) modulo 2 == 1 ) % ECN bits = ECT(1) or CE
7: lq.enqueue(pkt)
8: else % ECN bits = not-ECT or ECT(0)
9: cq.enqueue(pkt)
10: }
11: }
Figure 2: Example Enqueue Pseudocode for DualQ Coupled PI2 AQM
1: dualpi2_dequeue(lq, cq, pkt) { % Couples L4S & Classic queues
2: while ( lq.len() + cq.len() > 0 )
3: if ( lq.time() + tshift >= cq.time() ) { % time-shifted FIFO
4: lq.dequeue(pkt)
5: if ( ((pkt.time() > T_time) % step marking ...
6: AND (lq.len > T_len))
7: OR (p_L > rand()) ) % ...or linear marking
8: mark(pkt)
9: } else {
10: cq.dequeue(pkt)
11: if ( p_C > rand() ) { % probability p^2
12: if ( ecn(pkt) == 0 ) { % if ECN field = not-ECT
13: drop(pkt) % squared drop
14: continue % continue to the top of the while loop
15: }
16: mark(pkt) % squared mark
17: }
18: }
19: return(pkt) % return the packet and stop
20: }
21: return(NULL) % no packet to dequeue
22: }
Figure 3: Example Dequeue Pseudocode for DualQ Coupled PI2 AQM
1: dualpi2_update(lq, cq, target) { % Update p every Tupdate
2: curq = cq.time() % use queuing time of first-in Classic packet
3: p = p + alpha_U * (curq - target) + beta_U * (curq - prevq)
4a: p_L = p * k % L4S prob = base prob * coupling factor
4b: p_C = p^2 % Classic prob = (base prob)^2
5: prevq = curq
6: }
Figure 4: Example PI-Update Pseudocode for DualQ Coupled PI2 AQM
When packets arrive, first a common queue limit is checked as shown When packets arrive, first a common queue limit is checked as shown
in line 3 of the enqueuing pseudocode in Figure 1. Note that the in line 3 of the enqueuing pseudocode in Figure 2. Note that the
limit is deliberately tested before enqueue to avoid any bias against limit is deliberately tested before enqueue to avoid any bias against
larger packets (so the actual buffer has to be one packet larger than larger packets (so the actual buffer has to be one packet larger than
limit). If limit is not exceeded, the packet will be classified and limit). If limit is not exceeded, the packet will be classified and
enqueued to the Classic or L4S queue dependent on the least enqueued to the Classic or L4S queue dependent on the least
significant bit of the ECN field in the IP header (line 6). Packets significant bit of the ECN field in the IP header (line 6). Packets
with a codepoint having an LSB of 0 (Not-ECT and ECT(0)) will be with a codepoint having an LSB of 0 (Not-ECT and ECT(0)) will be
enqueued in the Classic queue. Otherwise, ECT(1) and CE packets will enqueued in the Classic queue. Otherwise, ECT(1) and CE packets will
be enqueued in the L4S queue. be enqueued in the L4S queue. Optional packet classification
flexibility is omitted for brevity.
The pseudocode in Figure 2 summarises the per packet dequeue The dequeue pseudocode schedules one packet for dequeuing (or zero if
implementation of the DualPI2 code. Line 3 implements the time- the queue is empty). It also makes all the AQM decisions on dropping
and marking. It is contained within a large while loop so that if it
decides to drop a packet, it will continue until it selects a packet
to schedule. Line 3 of the dequeue pseudocode implements time-
shifted FIFO scheduling. It takes the packet that waited the shifted FIFO scheduling. It takes the packet that waited the
longest, biased by a time-shift of tshift for the Classic traffic. longest, biased against the Classic traffic by a time-shift of
If an L4S packet is scheduled, lines 5 and 6 mark the packet if tshift.
either the L4S threshold T is exceeded, or if a random marking
decision is drawn according to k times the probability p (maintained o If an L4S packet is scheduled, lines 5 to 8 mark the packet if
by the dualpi2_update() function discussed below). The coupling either the L4S threshold is exceeded, or if a random marking
factor is applied here for determining the L4S marking probability so decision is drawn according to k times the probability p
that Classic TCP control is independent from the L4S coupling factor. (maintained by the dualpi2_update() function discussed below).
If a Classic packet is scheduled, lines 10 to 16 drop or mark the The L4S threshold is usually in units of time (default T_time = 1
packet based on 2 random decisions resulting in the squared ms). However, on slow links the packet serialization time can
probability p^2 (hence the name PI2 for Classic traffic). Note that approach the threshold T_time, so line 6 sets a floor of 2 MTU to
p is not reduced here by the factor k, as p has already been the threshold.
multiplied by the factor k when it was used to mark the L4S traffic.
The coupling factor gives Classic TCP and DCTCP traffic equal o If a Classic packet is scheduled, lines 10 to 17 drop or mark the
throughput; Because L4S marking is factored up by k, the dynamic gain packet based on the squared probability p_C = p^2 (hence the name
parameters alpha and beta also have to be factored up by k for the PI2 for Classic traffic).
L4S queue, which is necessary to ensure that Classic TCP and DCTCP
controls have the same stability.
The probability p is kept up to date by the core PI algorithm in The probability p is kept up to date by the core PI algorithm in
Figure 3 which is executed every Tupdate ([I-D.ietf-aqm-pie] now Figure 4 which is executed every Tupdate ([RFC8033] now recommends
recommends 16ms, but in our testing so far we have used the earlier 16ms). The algorithm centres on line 3, which is a classical
recommendation of 32ms). Note that p solely depends on the queuing Proportional-Integral (PI) controller that alters p dependent on a)
time in the Classic queue. In line 2, the current queuing delay is the error between the current queuing delay (curq) and the target
evaluated by inspecting the timestamp of the next packet to schedule queuing delay (target) as defined in [RFC8033] and b) the change in
in the Classic queue. The function cq.time() subtracts the time queuing delay since the last sample. The name 'PI' represents the
stamped at enqueue from the current time and implicitly takes the fact that the second factor is proportional to load while the first
current queuing delay as 0 if the queue is empty. Line 3 and 4 only is the integral of the load (so it removes any standing queue). In
need to be executed when the configuration parameters are changed. corner cases, p can overflow the range [0,1] so the resulting value
Alpha and beta in Hz are gain factors per 1 second. If a briefer of p has to be bounded (omitted from the pseudocode).
update time is configured, alpha_U and beta_U (_U = per Tupdate) also
have to be reduced, to ensure that the same response is given over
time. As such, a smaller Tupdate will only result in a response with
smaller and finer steps, not a more aggressive response. The new
probability is calculated in line 5, where target is the target
queuing delay, as defined in [I-D.ietf-aqm-pie]. In corner cases, p
can overflow the range [0,1] so the resulting value of p has to be
bounded (omitted from the pseudocode). Unlike PIE, alpha_U and
beta_U are not tuned dependent on p, every Tupdate. Instead, in PI2
alpha_U and beta_U can be constants because the squaring applied to
Classic traffic tunes them inherently, as explained in [PI216].
In our experiments so far (building on experiments with PIE) on Alpha_U and beta_U are gain factors chosen based on stability
broadband access links ranging from 4 Mb/s to 200 Mb/s with base RTTs analysis to convert changes in queueing delay into changes in
from 5 ms to 100 ms, PI2 achieves good results with the following probability. They are therefore in units of 'per second of delay' or
parameters: Hz. The suffix '_U' represents 'per update time' (Tupdate). If a
briefer update time is configured, alpha_U and beta_U also have to be
reduced to ensure that the same response is given over time, so that
a smaller Tupdate will only result in a response with finer steps,
not a more aggressive response. Therefore, alpha_U and beta_U are
derived from alpha and beta, which each represent a 'gain factor per
second of update time', so they can be configured independently of
the update time (see lines 7 and 8 of Figure 1).
tshift = 40ms Unlike in PIE, alpha_U and beta_U do not need to be tuned every
T = max(1ms, serialization time of 2 MTU) Tupdate dependent on p. Instead, in PI2 alpha_U and beta_U can be
constants because the squaring applied to Classic traffic tunes them
inherently, as explained in [PI216].
target = 20ms Note that p solely depends on the queuing time in the Classic queue.
In line 2, the current queuing delay is evaluated by inspecting the
timestamp of the next packet to schedule in the Classic queue. The
function cq.time() subtracts the time stamped at enqueue from the
current time and implicitly takes the current queuing delay as 0 if
the queue is empty.
Tupdate = 32ms Because the L4S marking probability (p_L) is factored up by k, the
dynamic gain parameters alpha and beta are also inherently factored
up by k for the L4S queue, which is necessary to ensure that Classic
TCP and DCTCP controls have the same stability. So, if alpha is 10
Hz^2, the effective gain factor for the L4S queue is k*alpha, which
is 20 Hz^2 with the default coupling factor of k=2.
k = 2 A.2. Pass #2: Overload Details
alpha = 10Hz (alpha*k = 20Hz for L4S) Figure 5 repeats the dequeue function of Figure 3, but with overload
details added. Similarly Figure 6 repeats the core PI algorithm of
Figure 4 with overload details added. The initialization and enqueue
functions are unchanged.
beta = 100Hz (beta*k = 200Hz for L4S) In line 13 of the initialization function (Figure 1), the default
maximum Classic drop probability p_Cmax = 1/4 or 25%. This is the
point at which it is deemed that the Classic queue has become
persistently overloaded, so it switches to using solely drop, even
for ECN-capable packets. This protects the queue against any
unresponsive traffic that falsely claims that it is responsive to ECN
marking, as required by [RFC3168] and [RFC7567].
Line 14 translates this into a maximum L4S marking probability
(p_Lmax) by rearranging Equation (1). With the default coupling
factor of k=2, this translates to a maximum L4S marking probability
of 1 or 100%. This is intended to ensure that the L4S queue starts to
introduce dropping once marking saturates and can rise no further.
The 'TCP Prague' requirements [I-D.ietf-tsvwg-ecn-l4s-id] require
that, when an L4S congestion control detects a drop, it falls back to
a response that coexists with 'Classic' TCP. So it is correct that
the L4S queue drops packets proportional to p^2, as if they are
Classic packets.
Both these switch-overs are triggered by the tests for overload
introduced in lines 4b and 12b of the dequeue function (Figure 5).
Lines 8c to 8g drop L4S packets with probability p^2 by comparing p
against two random numbers. Lines 8h to 8i mark the remaining
packets with probability p_L.
Lines 2c to 2d in the core PI algorithm (Figure 6) deal with overload
of the L4S queue when there is no Classic traffic. This is
necessary, because the core PI algorithm maintains the right
probability of drop to regulate overload, but it depends on the
length of the Classic queue. If there is no Classic queue the naive
algorithm in Figure 4 drops nothing, even if the L4S queue is
overloaded - so tail drop takes over (lines 3 and 4 of Figure 2).
If the test at line 2a finds that the Classic queue is empty, line 2d
measures the current queue delay using the L4S queue not the Classic
queue, While the L4S queue is not overloaded, its delay will always
be tiny compared to the target Classic queue delay. So p_L will be
driven to zero, and the L4S queue will naturally be governed solely
by threshold marking (lines 5 and 6 of the dequeue algorithm in
Figure 5). But, if unresponsive L4S source(s) cause overload, the
DualQ transitions smoothly to L4S marking based on the PI algorithm.
And as overload increases, it naturally transitions from marking to
dropping by the switch-over mechanism already described.
1: dualpi2_dequeue(lq, cq) { % Couples L4S & Classic queues, lq & cq
2: while ( lq.len() + cq.len() > 0 )
3: if ( lq.time() + tshift >= cq.time() ) { % time-shifted FIFO
4a: lq.dequeue(pkt)
4b: if ( p_L < p_Lmax ) { % Check for overload saturation
5: if ( ((pkt.time() > T_time) % step marking ...
6: AND (lq.len > T_len))
7: OR (p_L > rand()) ) % ...or linear marking
8a: mark(pkt)
8b: } else { % overload saturation
8c: if ( p > max(rand(), rand()) ) { % probability p^2
8e: drop(pkt) % revert to Classic drop due to overload
8f: continue % continue to the top of the while loop
8g: }
8h: if ( p_L > rand() ) % probability p_L = k * p
8i: mark(pkt) % linear marking of remaining packets
8j: }
9: } else {
10: cq.dequeue(pkt)
11: if ( p > max(rand(), rand()) ) { % probability p^2
12a: if ( (ecn(pkt) == 0) % ECN field = not-ECT
12b: OR (p_L >= p_Lmax) ) { % Overload disables ECN
13: drop(pkt) % squared drop, redo loop
14: continue % continue to the top of the while loop
15: }
16: mark(pkt) % squared mark
17: }
18: }
19: return(pkt) % return the packet and stop
20: }
21: return(NULL) % no packet to dequeue
22: }
Figure 5: Example Dequeue Pseudocode for DualQ Coupled PI2 AQM
(Including Integer Arithmetic and Overload Code)
1: dualpi2_update(lq, cq, target) { % Update p every Tupdate
2a: if ( cq.len() > 0 )
2b: curq = cq.time() %use queuing time of first-in Classic packet
2c: else % Classic queue empty
2d: curq = lq.time() % use queuing time of first-in L4S packet
3: p = p + alpha_U * (curq - target) + beta_U * (curq - prevq)
4: p_L = p * k % L4S prob = base prob * coupling factor
5: prevq = curq
6: }
Figure 6: Example PI-Update Pseudocode for DualQ Coupled PI2 AQM
(Including Overload Code)
Appendix B. Example DualQ Coupled Curvy RED Algorithm Appendix B. Example DualQ Coupled Curvy RED Algorithm
As another example, the pseudocode below gives the Curvy RED based As another example of a DualQ Coupled AQM algorithm, the pseudocode
DualQ Coupled AQM algorithm we used and tested. Although we designed below gives the Curvy RED based algorithm we used and tested.
the AQM to be efficient in integer arithmetic, to aid understanding Although we designed the AQM to be efficient in integer arithmetic,
it is first given using real-number arithmetic. Then, one possible to aid understanding it is first given using real-number arithmetic.
optimization for integer arithmetic is given, also in pseudocode. To Then, one possible optimization for integer arithmetic is given, also
aid comparison, the line numbers are kept in step between the two by in pseudocode. To aid comparison, the line numbers are kept in step
using letter suffixes where the longer code needs extra lines. between the two by using letter suffixes where the longer code needs
extra lines.
1: dualq_dequeue(lq, cq) { % Couples L4S & Classic queues, lq & cq 1: dualq_dequeue(lq, cq) { % Couples L4S & Classic queues, lq & cq
2: if ( lq.dequeue(pkt) ) { 2: if ( lq.dequeue(pkt) ) {
3a: p_L = cq.sec() / 2^S_L 3a: p_L = cq.sec() / 2^S_L
3b: if ( lq.byt() > T ) 3b: if ( lq.byt() > T )
3c: mark(pkt) 3c: mark(pkt)
3d: elif ( p_L > maxrand(U) ) 3d: elif ( p_L > maxrand(U) )
4: mark(pkt) 4: mark(pkt)
5: return(pkt) % return the packet and stop here 5: return(pkt) % return the packet and stop here
6: } 6: }
skipping to change at page 18, line 33 skipping to change at page 21, line 44
14: return(NULL) % no packet to dequeue 14: return(NULL) % no packet to dequeue
15: } 15: }
16: maxrand(u) { % return the max of u random numbers 16: maxrand(u) { % return the max of u random numbers
17: maxr=0 17: maxr=0
18: while (u-- > 0) 18: while (u-- > 0)
19: maxr = max(maxr, rand()) % 0 <= rand() < 1 19: maxr = max(maxr, rand()) % 0 <= rand() < 1
20: return(maxr) 20: return(maxr)
21: } 21: }
Figure 4: Example Dequeue Pseudocode for DualQ Coupled Curvy RED AQM Figure 7: Example Dequeue Pseudocode for DualQ Coupled Curvy RED AQM
Packet classification code is not shown, as it is no different from Packet classification code is not shown, as it is no different from
Figure 1. Potential classification schemes are discussed in Figure 2. Classic ECN handling is not shown. Potential
Section 2. Overload protection code will be included in a future classification schemes are discussed in Section 2. The Curvy RED
draft {ToDo}. algorithm has not been maintained to the same degree as the DualPI2
algorithm. Some ideas used in DualPI2 could be translated into Curvy
RED, such as i) the time-shifted FIFO scheduler ii) the time-based
L4S threshold; iii) overload protection. {ToDo}.
At the outer level, the structure of dualq_dequeue() implements At the outer level, the structure of dualq_dequeue() implements
strict priority scheduling. The code is written assuming the AQM is strict priority scheduling. The code is written assuming the AQM is
applied on dequeue (Note 1) . Every time dualq_dequeue() is called, applied on dequeue (Note 1) . Every time dualq_dequeue() is called,
the if-block in lines 2-6 determines whether there is an L4S packet the if-block in lines 2-6 determines whether there is an L4S packet
to dequeue by calling lq.dequeue(pkt), and otherwise the while-block to dequeue by calling lq.dequeue(pkt), and otherwise the while-block
in lines 7-13 determines whether there is a Classic packet to in lines 7-13 determines whether there is a Classic packet to
dequeue, by calling cq.dequeue(pkt). (Note 2) dequeue, by calling cq.dequeue(pkt). (Note 2)
In the lower priority Classic queue, a while loop is used so that, if In the lower priority Classic queue, a while loop is used so that, if
skipping to change at page 19, line 38 skipping to change at page 22, line 51
Specifically, in line 3a the marking probability p_L is set to the Specifically, in line 3a the marking probability p_L is set to the
Classic queueing time qc.sec() in seconds divided by the L4S Classic queueing time qc.sec() in seconds divided by the L4S
scaling parameter 2^S_L, which represents the queuing time (in scaling parameter 2^S_L, which represents the queuing time (in
seconds) at which marking probability would hit 100%. Then in line seconds) at which marking probability would hit 100%. Then in line
3d (if U=1) the result is compared with a uniformly distributed 3d (if U=1) the result is compared with a uniformly distributed
random number between 0 and 1, which ensures that marking random number between 0 and 1, which ensures that marking
probability will linearly increase with queueing time. The probability will linearly increase with queueing time. The
scaling parameter is expressed as a power of 2 so that division scaling parameter is expressed as a power of 2 so that division
can be implemented as a right bit-shift (>>) in line 3 of the can be implemented as a right bit-shift (>>) in line 3 of the
integer variant of the pseudocode (Figure 5). integer variant of the pseudocode (Figure 8).
Classic: If the test at line 7 determines that there is at least one Classic: If the test at line 7 determines that there is at least one
Classic packet to dequeue, the test at line 9b determines whether Classic packet to dequeue, the test at line 9b determines whether
to drop it. But before that, line 8b updates Q_C, which is an to drop it. But before that, line 8b updates Q_C, which is an
exponentially weighted moving average (Note 5) of the queuing time exponentially weighted moving average (Note 5) of the queuing time
in the Classic queue, where pkt.sec() is the instantaneous in the Classic queue, where pkt.sec() is the instantaneous
queueing time of the current Classic packet and alpha is the EWMA queueing time of the current Classic packet and alpha is the EWMA
constant for the classic queue. In line 8a, alpha is represented constant for the classic queue. In line 8a, alpha is represented
as an integer power of 2, so that in line 8 of the integer code as an integer power of 2, so that in line 8 of the integer code
the division needed to weight the moving average can be the division needed to weight the moving average can be
skipping to change at page 21, line 5 skipping to change at page 24, line 18
Classic: Classic:
S_C : The scaling factor of the dropping function scales Classic S_C : The scaling factor of the dropping function scales Classic
queuing times in the range [0, 2^(S_C)] seconds into a dropping queuing times in the range [0, 2^(S_C)] seconds into a dropping
probability in the range [0,1]. To make division efficient, it probability in the range [0,1]. To make division efficient, it
is constrained to be an integer power of two; is constrained to be an integer power of two;
f_C : To smooth the queuing time of the Classic queue and make f_C : To smooth the queuing time of the Classic queue and make
multiplication efficient, we use a negative integer power of multiplication efficient, we use a negative integer power of
two for the dimensionless EWMA constant, which we define as two for the dimensionless EWMA constant, which we define as
2^(-f_C). alpha = 2^(-f_C).
L4S : L4S :
S_L (and k): As for the Classic queue, the scaling factor of the S_L (and k'): As for the Classic queue, the scaling factor of
L4S marking function scales Classic queueing times in the range the L4S marking function scales Classic queueing times in the
[0, 2^(S_L)] seconds into a probability in the range [0,1]. range [0, 2^(S_L)] seconds into a probability in the range
Note that S_L = S_C + k, where k is the coupling between the [0,1]. Note that S_L = S_C + k', where k' is the coupling
queues (Section 2.1). So S_L and k count as only one between the queues. So S_L and k' count as only one parameter;
parameter; k' is related to k in Equation (1) (Section 2.1) by k=2^k',
where both k and k' are constants. Then implementations can
avoid costly division by shifting p_L by k' bits to the right.
T : The queue size in bytes at which step threshold marking T : The queue size in bytes at which step threshold marking
starts in the L4S queue. starts in the L4S queue.
{ToDo: These are the raw parameters used within the algorithm. A {ToDo: These are the raw parameters used within the algorithm. A
configuration front-end could accept more meaningful parameters and configuration front-end could accept more meaningful parameters and
convert them into these raw parameters.} convert them into these raw parameters.}
From our experiments so far, recommended values for these parameters From our experiments so far, recommended values for these parameters
are: S_C = -1; f_C = 5; T = 5 * MTU for the range of base RTTs are: S_C = -1; f_C = 5; T = 5 * MTU for the range of base RTTs
skipping to change at page 22, line 26 skipping to change at page 25, line 40
7: while ( cq.dequeue(pkt) ) { 7: while ( cq.dequeue(pkt) ) {
8: Q_C += (pkt.ns() - Q_C) >> f_C % Classic Q EWMA 8: Q_C += (pkt.ns() - Q_C) >> f_C % Classic Q EWMA
9: if ( (Q_C >> (S_C-2) ) > maxrand(2*U) ) 9: if ( (Q_C >> (S_C-2) ) > maxrand(2*U) )
10: drop(pkt) % Squared drop, redo loop 10: drop(pkt) % Squared drop, redo loop
11: else 11: else
12: return(pkt) % return the packet and stop here 12: return(pkt) % return the packet and stop here
13: } 13: }
14: return(NULL) % no packet to dequeue 14: return(NULL) % no packet to dequeue
15: } 15: }
Figure 5: Optimised Example Dequeue Pseudocode for Coupled DualQ AQM Figure 8: Optimised Example Dequeue Pseudocode for Coupled DualQ AQM
using Integer Arithmetic using Integer Arithmetic
Notes: Notes:
1. The drain rate of the queue can vary if it is scheduled relative 1. The drain rate of the queue can vary if it is scheduled relative
to other queues, or to cater for fluctuations in a wireless to other queues, or to cater for fluctuations in a wireless
medium. To auto-adjust to changes in drain rate, the queue must medium. To auto-adjust to changes in drain rate, the queue must
be measured in time, not bytes or packets [CoDel]. In our Linux be measured in time, not bytes or packets [CoDel]. In our Linux
implementation, it was easiest to measure queuing time at implementation, it was easiest to measure queuing time at
dequeue. Queuing time can be estimated when a packet is enqueued dequeue. Queuing time can be estimated when a packet is enqueued
skipping to change at page 23, line 21 skipping to change at page 26, line 34
adaptive smoothing methods could be valid and it might be adaptive smoothing methods could be valid and it might be
appropriate to decrease the EWMA faster than it increases. appropriate to decrease the EWMA faster than it increases.
6. In practice at line 10 the Classic queue would probably test for 6. In practice at line 10 the Classic queue would probably test for
ECN capability on the packet to determine whether to drop or mark ECN capability on the packet to determine whether to drop or mark
the packet. However, for brevity such detail is omitted. All the packet. However, for brevity such detail is omitted. All
packets classified into the L4S queue have to be ECN-capable, so packets classified into the L4S queue have to be ECN-capable, so
no dropping logic is necessary at line 3. Nonetheless, L4S no dropping logic is necessary at line 3. Nonetheless, L4S
packets could be dropped by overload code (see Section 4.1). packets could be dropped by overload code (see Section 4.1).
7. In the integer variant of the pseudocode (Figure 5) real numbers 7. In the integer variant of the pseudocode (Figure 8) real numbers
are all represented as integers scaled up by 2^32. In lines 3 & are all represented as integers scaled up by 2^32. In lines 3 &
9 the function maxrand() is arranged to return an integer in the 9 the function maxrand() is arranged to return an integer in the
range 0 <= maxrand() < 2^32. Queuing times are also scaled up by range 0 <= maxrand() < 2^32. Queuing times are also scaled up by
2^32, but in two stages: i) In lines 3 and 8 queuing times 2^32, but in two stages: i) In lines 3 and 8 queuing times
cq.ns() and pkt.ns() are returned in integer nanoseconds, making cq.ns() and pkt.ns() are returned in integer nanoseconds, making
the values about 2^30 times larger than when the units were the values about 2^30 times larger than when the units were
seconds, ii) then in lines 3 and 9 an adjustment of -2 to the seconds, ii) then in lines 3 and 9 an adjustment of -2 to the
right bit-shift multiplies the result by 2^2, to complete the right bit-shift multiplies the result by 2^2, to complete the
scaling by 2^32. scaling by 2^32.
skipping to change at page 23, line 33 skipping to change at page 27, line 4
9 the function maxrand() is arranged to return an integer in the 9 the function maxrand() is arranged to return an integer in the
range 0 <= maxrand() < 2^32. Queuing times are also scaled up by range 0 <= maxrand() < 2^32. Queuing times are also scaled up by
2^32, but in two stages: i) In lines 3 and 8 queuing times 2^32, but in two stages: i) In lines 3 and 8 queuing times
cq.ns() and pkt.ns() are returned in integer nanoseconds, making cq.ns() and pkt.ns() are returned in integer nanoseconds, making
the values about 2^30 times larger than when the units were the values about 2^30 times larger than when the units were
seconds, ii) then in lines 3 and 9 an adjustment of -2 to the seconds, ii) then in lines 3 and 9 an adjustment of -2 to the
right bit-shift multiplies the result by 2^2, to complete the right bit-shift multiplies the result by 2^2, to complete the
scaling by 2^32. scaling by 2^32.
Appendix C. Guidance on Controlling Throughput Equivalence Appendix C. Guidance on Controlling Throughput Equivalence
+---------------+------+-------+ +---------------+------+-------+
| RTT_C / RTT_L | Reno | Cubic | | RTT_C / RTT_L | Reno | Cubic |
+---------------+------+-------+ +---------------+------+-------+
| 1 | k=1 | k=0 | | 1 | k'=1 | k'=0 |
| 2 | k=2 | k=1 | | 2 | k'=2 | k'=1 |
| 3 | k=2 | k=2 | | 3 | k'=2 | k'=2 |
| 4 | k=3 | k=2 | | 4 | k'=3 | k'=2 |
| 5 | k=3 | k=3 | | 5 | k'=3 | k'=3 |
+---------------+------+-------+ +---------------+------+-------+
Table 1: Value of k for which DCTCP throughput is roughly the same as Table 1: Value of k' for which DCTCP throughput is roughly the same
Reno or Cubic, for some example RTT ratios as Reno or Cubic, for some example RTT ratios
k' is related to k in Equation (1) (Section 2.1) by k=2^k'.
To determine the appropriate policy, the operator first has to judge To determine the appropriate policy, the operator first has to judge
whether it wants DCTCP flows to have roughly equal throughput with whether it wants DCTCP flows to have roughly equal throughput with
Reno or with Cubic (because, even in its Reno-compatibility mode, Reno or with Cubic (because, even in its Reno-compatibility mode,
Cubic is about 1.4 times more aggressive than Reno). Then the Cubic is about 1.4 times more aggressive than Reno). Then the
operator needs to decide at what ratio of RTTs it wants DCTCP and operator needs to decide at what ratio of RTTs it wants DCTCP and
Classic flows to have roughly equal throughput. For example choosing Classic flows to have roughly equal throughput. For example choosing
the recommended value of k=0 will make DCTCP throughput roughly the k'=0 (equivalent to k=1) will make DCTCP throughput roughly the same
same as Cubic, _if their RTTs are the same_. as Cubic, _if their RTTs are the same_.
However, even if the base RTTs are the same, the actual RTTs are However, even if the base RTTs are the same, the actual RTTs are
unlikely to be the same, because Classic (Cubic or Reno) traffic unlikely to be the same, because Classic (Cubic or Reno) traffic
needs a large queue to avoid under-utilization and excess drop, needs a large queue to avoid under-utilization and excess drop,
whereas L4S (DCTCP) does not. The operator might still choose this whereas L4S (DCTCP) does not. The operator might still choose this
policy if it judges that DCTCP throughput should be rewarded for policy if it judges that DCTCP throughput should be rewarded for
keeping its own queue short. keeping its own queue short.
On the other hand, the operator will choose one of the higher values On the other hand, the operator will choose one of the higher values
for k, if it wants to slow DCTCP down to roughly the same throughput for k', if it wants to slow DCTCP down to roughly the same throughput
as Classic flows, to compensate for Classic flows slowing themselves as Classic flows, to compensate for Classic flows slowing themselves
down by causing themselves extra queuing delay. down by causing themselves extra queuing delay.
The values for k in the table are derived from the formulae, which The values for k' in the table are derived from the formulae, which
was developed in [DCttH15]: was developed in [DCttH15]:
2^k = 1.64 (RTT_reno / RTT_dc) (2) 2^k' = 1.64 (RTT_reno / RTT_dc) (2)
2^k = 1.19 (RTT_cubic / RTT_dc ) (3) 2^k' = 1.19 (RTT_cubic / RTT_dc ) (3)
For localized traffic from a particular ISP's data centre, we used For localized traffic from a particular ISP's data centre, we used
the measured RTTs to calculate that a value of k=3 would achieve the measured RTTs to calculate that a value of k'=3 (equivalant to
throughput equivalence, and our experiments verified the formula very k=8) would achieve throughput equivalence, and our experiments
closely. verified the formula very closely.
For a typical mix of RTTs from local data centres and across the
general Internet, a value of k'=1 (equivalent to k=2) is recommended
as a good workable compromise.
Authors' Addresses Authors' Addresses
Koen De Schepper Koen De Schepper
Nokia Bell Labs Nokia Bell Labs
Antwerp Antwerp
Belgium Belgium
Email: koen.de_schepper@nokia.com Email: koen.de_schepper@nokia.com
URI: https://www.bell-labs.com/usr/koen.de_schepper URI: https://www.bell-labs.com/usr/koen.de_schepper
 End of changes. 54 change blocks. 
183 lines changed or deleted 357 lines changed or added

This html diff was produced by rfcdiff 1.45. The latest version is available from http://tools.ietf.org/tools/rfcdiff/