< draft-ietf-tcpm-rfc8312bis-05.txt   draft-ietf-tcpm-rfc8312bis-06.txt >
TCPM L. Xu TCPM L. Xu
Internet-Draft UNL Internet-Draft UNL
Obsoletes: 8312 (if approved) S. Ha Obsoletes: 8312 (if approved) S. Ha
Updates: 5681 (if approved) Colorado Updates: 5681 (if approved) Colorado
Intended status: Standards Track I. Rhee Intended status: Standards Track I. Rhee
Expires: 28 April 2022 Bowery Expires: 30 July 2022 Bowery
V. Goel V. Goel
Apple Inc. Apple Inc.
L. Eggert, Ed. L. Eggert, Ed.
NetApp NetApp
25 October 2021 26 January 2022
CUBIC for Fast and Long-Distance Networks CUBIC for Fast and Long-Distance Networks
draft-ietf-tcpm-rfc8312bis-05 draft-ietf-tcpm-rfc8312bis-06
Abstract Abstract
CUBIC is a standard TCP congestion control algorithm that uses a CUBIC is a standard TCP congestion control algorithm that uses a
cubic function instead of the linear window increase function on the cubic function instead of a linear congestion window increase
sender side to improve scalability and stability over fast and long- function to improve scalability and stability over fast and long-
distance networks. CUBIC has been adopted as the default TCP distance networks. CUBIC has been adopted as the default TCP
congestion control algorithm by the Linux, Windows, and Apple stacks. congestion control algorithm by the Linux, Windows, and Apple stacks.
This document updates the specification of CUBIC to include This document updates the specification of CUBIC to include
algorithmic improvements based on these implementations and recent algorithmic improvements based on these implementations and recent
academic work. Based on the extensive deployment experience with academic work. Based on the extensive deployment experience with
CUBIC, it also moves the specification to the Standards Track, CUBIC, it also moves the specification to the Standards Track,
obsoleting RFC 8312. This also requires updating RFC 5681, to allow obsoleting RFC 8312. This also requires updating RFC 5681, to allow
for CUBIC's occasionally more aggressive sending behavior. for CUBIC's occasionally more aggressive sending behavior.
Note to Readers About This Document
Discussion of this draft takes place on the TCPM working group This note is to be removed before publishing as an RFC.
Status information for this document may be found at
https://datatracker.ietf.org/doc/draft-ietf-tcpm-rfc8312bis/.
Discussion of this document takes place on the TCPM Working Group
mailing list (mailto:tcpm@ietf.org), which is archived at mailing list (mailto:tcpm@ietf.org), which is archived at
https://mailarchive.ietf.org/arch/browse/tcpm/. https://mailarchive.ietf.org/arch/browse/tcpm/.
Working Group information can be found at Source for this draft and an issue tracker can be found at
https://datatracker.ietf.org/wg/tcpm/; source code and issues list https://github.com/NTAP/rfc8312bis.
for this draft can be found at https://github.com/NTAP/rfc8312bis.
Note to the RFC Editor Note to the RFC Editor
xml2rfc currently renders <em></em> in the XML by surrounding the xml2rfc currently renders <em></em> in the XML by surrounding the
corresponding text with underscores. This is highly distracting; corresponding text with underscores. This is highly distracting;
please manually remove the underscores when doing the final edits to please manually remove the underscores when doing the final edits to
the text version of this document. the text version of this document.
(There is an issue open against xml2rfc to stop doing this in the (There is an issue open against xml2rfc to stop doing this in the
future: https://trac.tools.ietf.org/tools/xml2rfc/trac/ticket/596) future: https://trac.tools.ietf.org/tools/xml2rfc/trac/ticket/596)
Also, please manually change "Figure" to "Equation" for all artwork Also, please manually change "Figure" to "Equation" for all artwork
with anchors beginning with "eq" - xml2rfc doesn't seem to be able to with anchors beginning with "eq" - xml2rfc doesn't seem to be able to
skipping to change at page 2, line 27 skipping to change at page 2, line 31
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 https://datatracker.ietf.org/drafts/current/. Drafts is at https://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 28 April 2022. This Internet-Draft will expire on 30 July 2022.
Copyright Notice Copyright Notice
Copyright (c) 2021 IETF Trust and the persons identified as the Copyright (c) 2022 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 (https://trustee.ietf.org/ Provisions Relating to IETF Documents (https://trustee.ietf.org/
license-info) in effect on the date of publication of this document. license-info) in effect on the date of publication of this document.
Please review these documents carefully, as they describe your rights Please review these documents carefully, as they describe your rights
and restrictions with respect to this document. Code Components and restrictions with respect to this document. Code Components
extracted from this document must include Simplified BSD License text extracted from this document must include Revised BSD License text as
as described in Section 4.e of the Trust Legal Provisions and are described in Section 4.e of the Trust Legal Provisions and are
provided without warranty as described in the Simplified BSD License. provided without warranty as described in the Revised BSD License.
Table of Contents Table of Contents
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 4 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 4
2. Conventions . . . . . . . . . . . . . . . . . . . . . . . . . 5 2. Conventions . . . . . . . . . . . . . . . . . . . . . . . . . 5
3. Design Principles of CUBIC . . . . . . . . . . . . . . . . . 5 3. Design Principles of CUBIC . . . . . . . . . . . . . . . . . 5
3.1. Principle 1 for the CUBIC Increase Function . . . . . . . 5 3.1. Principle 1 for the CUBIC Increase Function . . . . . . . 6
3.2. Principle 2 for Reno-Friendliness . . . . . . . . . . . . 6 3.2. Principle 2 for Reno-Friendliness . . . . . . . . . . . . 6
3.3. Principle 3 for RTT Fairness . . . . . . . . . . . . . . 7 3.3. Principle 3 for RTT Fairness . . . . . . . . . . . . . . 7
3.4. Principle 4 for the CUBIC Decrease Factor . . . . . . . . 7 3.4. Principle 4 for the CUBIC Decrease Factor . . . . . . . . 7
4. CUBIC Congestion Control . . . . . . . . . . . . . . . . . . 8 4. CUBIC Congestion Control . . . . . . . . . . . . . . . . . . 8
4.1. Definitions . . . . . . . . . . . . . . . . . . . . . . . 8 4.1. Definitions . . . . . . . . . . . . . . . . . . . . . . . 8
4.1.1. Constants of Interest . . . . . . . . . . . . . . . . 8 4.1.1. Constants of Interest . . . . . . . . . . . . . . . . 8
4.1.2. Variables of Interest . . . . . . . . . . . . . . . . 8 4.1.2. Variables of Interest . . . . . . . . . . . . . . . . 8
4.2. Window Increase Function . . . . . . . . . . . . . . . . 9 4.2. Window Increase Function . . . . . . . . . . . . . . . . 9
4.3. Reno-Friendly Region . . . . . . . . . . . . . . . . . . 11 4.3. Reno-Friendly Region . . . . . . . . . . . . . . . . . . 11
4.4. Concave Region . . . . . . . . . . . . . . . . . . . . . 13 4.4. Concave Region . . . . . . . . . . . . . . . . . . . . . 13
4.5. Convex Region . . . . . . . . . . . . . . . . . . . . . . 13 4.5. Convex Region . . . . . . . . . . . . . . . . . . . . . . 13
4.6. Multiplicative Decrease . . . . . . . . . . . . . . . . . 14 4.6. Multiplicative Decrease . . . . . . . . . . . . . . . . . 14
4.7. Fast Convergence . . . . . . . . . . . . . . . . . . . . 15 4.7. Fast Convergence . . . . . . . . . . . . . . . . . . . . 15
4.8. Timeout . . . . . . . . . . . . . . . . . . . . . . . . . 15 4.8. Timeout . . . . . . . . . . . . . . . . . . . . . . . . . 16
4.9. Spurious Congestion Events . . . . . . . . . . . . . . . 16 4.9. Spurious Congestion Events . . . . . . . . . . . . . . . 16
4.10. Slow Start . . . . . . . . . . . . . . . . . . . . . . . 17 4.9.1. Spurious timeout . . . . . . . . . . . . . . . . . . 16
5. Discussion . . . . . . . . . . . . . . . . . . . . . . . . . 17 4.9.2. Spurious loss detected by acknowledgements . . . . . 17
5.1. Fairness to Reno . . . . . . . . . . . . . . . . . . . . 18 4.10. Slow Start . . . . . . . . . . . . . . . . . . . . . . . 18
5.2. Using Spare Capacity . . . . . . . . . . . . . . . . . . 20 5. Discussion . . . . . . . . . . . . . . . . . . . . . . . . . 18
5.3. Difficult Environments . . . . . . . . . . . . . . . . . 21 5.1. Fairness to Reno . . . . . . . . . . . . . . . . . . . . 19
5.4. Investigating a Range of Environments . . . . . . . . . . 21 5.2. Using Spare Capacity . . . . . . . . . . . . . . . . . . 21
5.5. Protection against Congestion Collapse . . . . . . . . . 22 5.3. Difficult Environments . . . . . . . . . . . . . . . . . 22
5.4. Investigating a Range of Environments . . . . . . . . . . 22
5.5. Protection against Congestion Collapse . . . . . . . . . 23
5.6. Fairness within the Alternative Congestion Control 5.6. Fairness within the Alternative Congestion Control
Algorithm . . . . . . . . . . . . . . . . . . . . . . . 22 Algorithm . . . . . . . . . . . . . . . . . . . . . . . 23
5.7. Performance with Misbehaving Nodes and Outside 5.7. Performance with Misbehaving Nodes and Outside
Attackers . . . . . . . . . . . . . . . . . . . . . . . 22 Attackers . . . . . . . . . . . . . . . . . . . . . . . 23
5.8. Behavior for Application-Limited Flows . . . . . . . . . 22 5.8. Behavior for Application-Limited Flows . . . . . . . . . 23
5.9. Responses to Sudden or Transient Events . . . . . . . . . 22 5.9. Responses to Sudden or Transient Events . . . . . . . . . 24
5.10. Incremental Deployment . . . . . . . . . . . . . . . . . 23 5.10. Incremental Deployment . . . . . . . . . . . . . . . . . 24
6. Security Considerations . . . . . . . . . . . . . . . . . . . 23 6. Security Considerations . . . . . . . . . . . . . . . . . . . 24
7. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 23 7. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 24
8. References . . . . . . . . . . . . . . . . . . . . . . . . . 23 8. References . . . . . . . . . . . . . . . . . . . . . . . . . 24
8.1. Normative References . . . . . . . . . . . . . . . . . . 23 8.1. Normative References . . . . . . . . . . . . . . . . . . 24
8.2. Informative References . . . . . . . . . . . . . . . . . 25 8.2. Informative References . . . . . . . . . . . . . . . . . 26
Appendix A. Acknowledgments . . . . . . . . . . . . . . . . . . 27 Appendix A. Acknowledgments . . . . . . . . . . . . . . . . . . 29
Appendix B. Evolution of CUBIC . . . . . . . . . . . . . . . . . 28 Appendix B. Evolution of CUBIC . . . . . . . . . . . . . . . . . 30
B.1. Since draft-ietf-tcpm-rfc8312bis-04 . . . . . . . . . . . 28 B.1. Since draft-ietf-tcpm-rfc8312bis-05 . . . . . . . . . . . 30
B.2. Since draft-ietf-tcpm-rfc8312bis-03 . . . . . . . . . . . 29 B.2. Since draft-ietf-tcpm-rfc8312bis-04 . . . . . . . . . . . 30
B.3. Since draft-ietf-tcpm-rfc8312bis-02 . . . . . . . . . . . 29 B.3. Since draft-ietf-tcpm-rfc8312bis-03 . . . . . . . . . . . 31
B.4. Since draft-ietf-tcpm-rfc8312bis-01 . . . . . . . . . . . 29 B.4. Since draft-ietf-tcpm-rfc8312bis-02 . . . . . . . . . . . 31
B.5. Since draft-ietf-tcpm-rfc8312bis-00 . . . . . . . . . . . 30 B.5. Since draft-ietf-tcpm-rfc8312bis-01 . . . . . . . . . . . 32
B.6. Since draft-eggert-tcpm-rfc8312bis-03 . . . . . . . . . . 30 B.6. Since draft-ietf-tcpm-rfc8312bis-00 . . . . . . . . . . . 32
B.7. Since draft-eggert-tcpm-rfc8312bis-02 . . . . . . . . . . 30 B.7. Since draft-eggert-tcpm-rfc8312bis-03 . . . . . . . . . . 32
B.8. Since draft-eggert-tcpm-rfc8312bis-01 . . . . . . . . . . 30 B.8. Since draft-eggert-tcpm-rfc8312bis-02 . . . . . . . . . . 32
B.9. Since draft-eggert-tcpm-rfc8312bis-00 . . . . . . . . . . 30 B.9. Since draft-eggert-tcpm-rfc8312bis-01 . . . . . . . . . . 32
B.10. Since RFC8312 . . . . . . . . . . . . . . . . . . . . . . 31 B.10. Since draft-eggert-tcpm-rfc8312bis-00 . . . . . . . . . . 33
B.11. Since the Original Paper . . . . . . . . . . . . . . . . 31 B.11. Since RFC8312 . . . . . . . . . . . . . . . . . . . . . . 33
Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 32 B.12. Since the Original Paper . . . . . . . . . . . . . . . . 34
Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 34
1. Introduction 1. Introduction
CUBIC has been adopted as the default TCP congestion control CUBIC has been adopted as the default TCP congestion control
algorithm in the Linux, Windows, and Apple stacks, and has been used algorithm in the Linux, Windows, and Apple stacks, and has been used
and deployed globally. Extensive, decade-long deployment experience and deployed globally. Extensive, decade-long deployment experience
in vastly different Internet scenarios has convincingly demonstrated in vastly different Internet scenarios has convincingly demonstrated
that CUBIC is safe for deployment on the global Internet and delivers that CUBIC is safe for deployment on the global Internet and delivers
substantial benefits over classical Reno congestion control substantial benefits over classical Reno congestion control
[RFC5681]. It is therefore to be regarded as the currently most [RFC5681]. It is therefore to be regarded as the currently most
widely deployed standard for TCP congestion control. CUBIC can also widely deployed standard for TCP congestion control. CUBIC can also
be used for other transport protocols such as QUIC [RFC9000] and SCTP be used for other transport protocols such as QUIC [RFC9000] and SCTP
[RFC4960] as a default congestion controller. [RFC4960] as a default congestion controller.
The design of CUBIC was motivated by the well-documented problem The design of CUBIC was motivated by the well-documented problem
classical Reno TCP has with low utilization over fast and long- classical Reno TCP has with low utilization over fast and long-
distance networks [K03][RFC3649]. This problem arises from a slow distance networks [K03][RFC3649]. This problem arises from a slow
increase of the congestion window following a congestion event in a increase of the congestion window following a congestion event in a
network with a large bandwidth-delay product (BDP). [HKLRX06] network with a large bandwidth-delay product (BDP). [HLRX07]
indicates that this problem is frequently observed even in the range indicates that this problem is frequently observed even in the range
of congestion window sizes over several hundreds of packets. This of congestion window sizes over several hundreds of packets. This
problem is equally applicable to all Reno-style standards and their problem is equally applicable to all Reno-style standards and their
variants, including TCP-Reno [RFC5681], TCP-NewReno variants, including TCP-Reno [RFC5681], TCP-NewReno
[RFC6582][RFC6675], SCTP [RFC4960], TFRC [RFC5348], and QUIC [RFC6582][RFC6675], SCTP [RFC4960], TFRC [RFC5348], and QUIC
congestion control [RFC9002], which use the same linear increase congestion control [RFC9002], which use the same linear increase
function for window growth. We refer to all Reno-style standards and function for window growth. We refer to all Reno-style standards and
their variants collectively as "Reno" below. their variants collectively as "Reno" below.
CUBIC, originally proposed in [HRX08], is a modification to the CUBIC, originally proposed in [HRX08], is a modification to the
skipping to change at page 6, line 20 skipping to change at page 6, line 29
of the congestion window. When CUBIC enters into congestion of the congestion window. When CUBIC enters into congestion
avoidance, it starts to increase the congestion window using the avoidance, it starts to increase the congestion window using the
concave profile of the cubic function. The cubic function is set to concave profile of the cubic function. The cubic function is set to
have its plateau at the remembered congestion window size, so that have its plateau at the remembered congestion window size, so that
the concave window increase continues until then. After that, the the concave window increase continues until then. After that, the
cubic function turns into a convex profile and the convex window cubic function turns into a convex profile and the convex window
increase begins. increase begins.
This style of window adjustment (concave and then convex) improves This style of window adjustment (concave and then convex) improves
the algorithm stability while maintaining high network utilization the algorithm stability while maintaining high network utilization
[CEHRX07]. This is because the window size remains almost constant, [CEHRX09]. This is because the window size remains almost constant,
forming a plateau around the remembered congestion window size of the forming a plateau around the remembered congestion window size of the
last congestion event, where network utilization is deemed highest. last congestion event, where network utilization is deemed highest.
Under steady state, most window size samples of CUBIC are close to Under steady state, most window size samples of CUBIC are close to
that remembered congestion window size, thus promoting high network that remembered congestion window size, thus promoting high network
utilization and stability. utilization and stability.
Note that congestion control algorithms that only use convex Note that congestion control algorithms that only use convex
functions to increase the congestion window size have their maximum functions to increase the congestion window size have their maximum
increments around the remembered congestion window size of the last increments around the remembered congestion window size of the last
congestion event, and thus introduce many packet bursts around the congestion event, and thus introduce many packet bursts around the
skipping to change at page 9, line 50 skipping to change at page 10, line 6
than the MSS. than the MSS.
4.2. Window Increase Function 4.2. Window Increase Function
CUBIC maintains the acknowledgment (ACK) clocking of Reno by CUBIC maintains the acknowledgment (ACK) clocking of Reno by
increasing the congestion window only at the reception of a new ACK. increasing the congestion window only at the reception of a new ACK.
It does not make any changes to the TCP Fast Recovery and Fast It does not make any changes to the TCP Fast Recovery and Fast
Retransmit algorithms [RFC6582][RFC6675]. Retransmit algorithms [RFC6582][RFC6675].
During congestion avoidance, after a congestion event is detected by During congestion avoidance, after a congestion event is detected by
mechanisms described in Section 3.1, CUBIC changes the window mechanisms described in Section 3.1, CUBIC uses a window increase
increase function of Reno. function different from Reno.
CUBIC uses the following window increase function: CUBIC uses the following window increase function:
3 3
W (t) = C * (t - K) + W W (t) = C * (t - K) + W
cubic max cubic max
Figure 1 Figure 1
where _t_ is the elapsed time in seconds from the beginning of the where _t_ is the elapsed time in seconds from the beginning of the
skipping to change at page 10, line 42 skipping to change at page 10, line 46
Figure 2 Figure 2
where _cwnd_start_ is the congestion window at the beginning of the where _cwnd_start_ is the congestion window at the beginning of the
current congestion avoidance stage. current congestion avoidance stage.
Upon receiving a new ACK during congestion avoidance, CUBIC computes Upon receiving a new ACK during congestion avoidance, CUBIC computes
the _target_ congestion window size after the next _RTT_ using the _target_ congestion window size after the next _RTT_ using
Figure 1 as follows, where _RTT_ is the smoothed round-trip time. Figure 1 as follows, where _RTT_ is the smoothed round-trip time.
The lower and upper bounds below ensure that CUBIC's congestion The lower and upper bounds below ensure that CUBIC's congestion
window increase rate is non-decreasing and is less than the increase window increase rate is non-decreasing and is less than the increase
rate of slow start. rate of slow start [SXEZ19].
/ /
| if W (t + RTT) < cwnd | if W (t + RTT) < cwnd
|cwnd cubic |cwnd cubic
| |
| |
| |
target = < if W (t + RTT) > 1.5 * cwnd target = < if W (t + RTT) > 1.5 * cwnd
|1.5 * cwnd cubic |1.5 * cwnd cubic
| |
| |
|W (t + RTT) |W (t + RTT)
| cubic otherwise | cubic otherwise
\ \
The elapsed time _t_ in Figure 1 MUST NOT include periods during The elapsed time _t_ in Figure 1 MUST NOT include periods during
which _cwnd_ has not been updated due to an application limit (see which _cwnd_ has not been updated due to application-limited behavior
Section 5.8). (see Section 5.8).
Depending on the value of the current congestion window size _cwnd_, Depending on the value of the current congestion window size _cwnd_,
CUBIC runs in three different regions: CUBIC runs in three different regions:
1. The Reno-friendly region, which ensures that CUBIC achieves at 1. The Reno-friendly region, which ensures that CUBIC achieves at
least the same throughput as Reno. least the same throughput as Reno.
2. The concave region, if CUBIC is not in the Reno-friendly region 2. The concave region, if CUBIC is not in the Reno-friendly region
and _cwnd_ is less than _W_max_. and _cwnd_ is less than _W_max_.
skipping to change at page 14, line 26 skipping to change at page 14, line 26
during fast recovery. during fast recovery.
In Figure 5, _flight_size_ is the amount of outstanding data in the In Figure 5, _flight_size_ is the amount of outstanding data in the
network, as defined in [RFC5681]. Note that a rate-limited network, as defined in [RFC5681]. Note that a rate-limited
application with idle periods or periods when unable to send at the application with idle periods or periods when unable to send at the
full rate permitted by _cwnd_ may easily encounter notable variations full rate permitted by _cwnd_ may easily encounter notable variations
in the volume of data sent from one RTT to another, resulting in in the volume of data sent from one RTT to another, resulting in
_flight_size_ that is significantly less than _cwnd_ on a congestion _flight_size_ that is significantly less than _cwnd_ on a congestion
event. This may decrease _cwnd_ to a much lower value than event. This may decrease _cwnd_ to a much lower value than
necessary. To avoid suboptimal performance with such applications, necessary. To avoid suboptimal performance with such applications,
some implementations of CUBIC use _cwnd_ instead of _flight_size_ to the mechanisms described in [RFC7661] can be used to mitigate this
calculate the new _ssthresh_ in Figure 5. Alternatively, the issue as it would allow using a value between _cwnd_ and
mechanisms described in [RFC7661] may also be adopted to mitigate _flight_size_ to calculate the new _ssthresh_ in Figure 5. Some
this issue. implementations of CUBIC use _cwnd_ instead of _flight_size_ when
calculating a new _ssthresh_ using Figure 5.
flight_size * β // new ssthresh flight_size * β // new ssthresh
ssthresh = cubic ssthresh = cubic
/max(ssthresh, 2) // reduction on packet loss, cwnd is at least 2 MSS /max(ssthresh, 2) // reduction on packet loss, cwnd is at least 2 MSS
| |
cwnd = < cwnd = <
|max(ssthresh, 1) // reduction on ECE, cwnd is at least 1 MSS |max(ssthresh, 1) // reduction on ECE, cwnd is at least 1 MSS
\ \
max(ssthresh, 2) // ssthresh is at least 2 MSS max(ssthresh, 2) // ssthresh is at least 2 MSS
ssthresh = ssthresh =
Figure 5 Figure 5
A side effect of setting β__cubic_ to a value bigger than 0.5 is A side effect of setting β__cubic_ to a value bigger than 0.5 is
slower convergence. We believe that while a more adaptive setting of slower convergence. We believe that while a more adaptive setting of
β__cubic_ could result in faster convergence, it will make the β__cubic_ could result in faster convergence, it will make the
analysis of CUBIC much harder. analysis of CUBIC much harder.
Note that CUBIC will continue to reduce _cwnd_ in response to Note that CUBIC MUST continue to reduce _cwnd_ in response to
congestion events due to ECN-Echo ACKs until it reaches a value of 1 congestion events due to ECN-Echo ACKs until it reaches a value of 1
MSS. If congestion persists, a sender with a _cwnd_ of 1 MSS needs MSS. If congestion events indicated by ECN-Echo ACKs persist, a
to reduce its sending rate even further. It can achieve that by sender with a _cwnd_ of 1 MSS MUST reduce its sending rate even
using a retransmission timer with exponential backoff, as described further. It can achieve that by using a retransmission timer with
in [RFC3168]. exponential backoff, as described in [RFC3168].
4.7. Fast Convergence 4.7. Fast Convergence
To improve convergence speed, CUBIC uses a heuristic. When a new To improve convergence speed, CUBIC uses a heuristic. When a new
flow joins the network, existing flows need to give up some of their flow joins the network, existing flows need to give up some of their
bandwidth to allow the new flow some room for growth, if the existing bandwidth to allow the new flow some room for growth, if the existing
flows have been using all the network bandwidth. To speed up this flows have been using all the network bandwidth. To speed up this
bandwidth release by existing flows, the following "Fast Convergence" bandwidth release by existing flows, the following "Fast Convergence"
mechanism SHOULD be implemented. mechanism SHOULD be implemented.
skipping to change at page 16, line 26 skipping to change at page 16, line 32
In cases where CUBIC reduces its congestion window in response to In cases where CUBIC reduces its congestion window in response to
having detected packet loss via duplicate ACKs or timeouts, there is having detected packet loss via duplicate ACKs or timeouts, there is
a possibility that the missing ACK would arrive after the congestion a possibility that the missing ACK would arrive after the congestion
window reduction and a corresponding packet retransmission. For window reduction and a corresponding packet retransmission. For
example, packet reordering could trigger this behavior. A high example, packet reordering could trigger this behavior. A high
degree of packet reordering could cause multiple congestion window degree of packet reordering could cause multiple congestion window
reduction events, where spurious losses are incorrectly interpreted reduction events, where spurious losses are incorrectly interpreted
as congestion signals, thus degrading CUBIC's performance as congestion signals, thus degrading CUBIC's performance
significantly. significantly.
When there is a congestion event, a CUBIC implementation SHOULD save For TCP, there are two types of spurious events - spurious timeouts
the current value of the following variables before the congestion and spurious fast retransmits. In case of QUIC, there are no
window reduction. spurious timeouts as the loss is only detected after receiving an
ACK.
4.9.1. Spurious timeout
An implementation MAY detect spurious timeouts based on the
mechanisms described in Forward RTO-Recovery [RFC5682]. Experimental
alternatives include Eifel [RFC3522]. When a spurious timeout is
detected, a TCP implementation MAY follow the response algorithm
described in [RFC4015] to restore the congestion control state and
adapt the retransmission timer to avoid further spurious timeouts.
4.9.2. Spurious loss detected by acknowledgements
Upon receiving an ACK, a TCP implementation MAY detect spurious
losses either using TCP Timestamps or via D-SACK[RFC2883].
Experimental alternatives include Eifel detection algorithm [RFC3522]
which uses TCP Timestamps and DSACK based detection [RFC3708] which
uses DSACK information. A QUIC implementation can easily determine a
spurious loss if a QUIC packet is acknowledged after it has been
marked as lost and the original data has been retransmitted with a
new QUIC packet.
In this section, we specify a simple response algorithm when a
spurious loss is detected by acknowledgements. Implementations would
need to carefully evaluate the impact of using this algorithm in
different environments that may experience sudden change in available
capacity (e.g., due to variable radio capacity, a routing change, or
a mobility event).
When a packet loss is detected via acknowledgements, a CUBIC
implementation MAY save the current value of the following variables
before the congestion window is reduced.
prior_cwnd = cwnd prior_cwnd = cwnd
prior_ssthresh = ssthresh prior_ssthresh = ssthresh
prior_W = W prior_W = W
max max max max
prior_K = K prior_K = K
prior_epoch = epoch prior_epoch = epoch
start start start start
prior_W_{est} = W prior_W_{est} = W
est est
CUBIC MAY implement an algorithm to detect spurious retransmissions, Once the previously declared packet loss is confirmed to be spurious,
such as Forward RTO-Recovery [RFC5682]. Experimental alternatives CUBIC MAY restore the original values of the above-mentioned
include DSACK [RFC3708] and Eifel [RFC3522]. Once a spurious variables as follows if the current _cwnd_ is lower than
congestion event is detected, CUBIC SHOULD restore the original _prior_cwnd_. Restoring the original values ensures that CUBIC's
values of above-mentioned variables as follows if the current _cwnd_ performance is similar to what it would be without spurious losses.
is lower than _prior_cwnd_. Restoring the original values ensures
that CUBIC's performance is similar to what it would be without
spurious losses.
\ \
cwnd = prior_cwnd | cwnd = prior_cwnd |
| |
ssthresh = prior_ssthresh | ssthresh = prior_ssthresh |
| |
W = prior_W | W = prior_W |
max max | max max |
>if cwnd < prior_cwnd >if cwnd < prior_cwnd
K = prior_K | K = prior_K |
skipping to change at page 17, line 33 skipping to change at page 18, line 33
CUBIC SHOULD continue to use the current and the most recent values CUBIC SHOULD continue to use the current and the most recent values
of these variables. of these variables.
4.10. Slow Start 4.10. Slow Start
CUBIC MUST employ a slow-start algorithm, when _cwnd_ is no more than CUBIC MUST employ a slow-start algorithm, when _cwnd_ is no more than
_ssthresh_. In general, CUBIC SHOULD use the HyStart++ slow start _ssthresh_. In general, CUBIC SHOULD use the HyStart++ slow start
algorithm [I-D.ietf-tcpm-hystartplusplus], or MAY use the Reno TCP algorithm [I-D.ietf-tcpm-hystartplusplus], or MAY use the Reno TCP
slow start algorithm [RFC5681] in the rare cases when HyStart++ is slow start algorithm [RFC5681] in the rare cases when HyStart++ is
not suitable. Experimental alternatives include hybrid slow start not suitable. Experimental alternatives include hybrid slow start
[HR08], a predecessor to HyStart++ that some CUBIC implementations [HR11], a predecessor to HyStart++ that some CUBIC implementations
have used as the default for the last decade, and limited slow start have used as the default for the last decade, and limited slow start
[RFC3742]. Whichever start-up algorithm is used, work might be [RFC3742]. Whichever start-up algorithm is used, work might be
needed to ensure that the end of slow start and the first needed to ensure that the end of slow start and the first
multiplicative decrease of congestion avoidance work well together. multiplicative decrease of congestion avoidance work well together.
When CUBIC uses HyStart++ [I-D.ietf-tcpm-hystartplusplus], it may When CUBIC uses HyStart++ [I-D.ietf-tcpm-hystartplusplus], it may
exit the first slow start without incurring any packet loss and thus exit the first slow start without incurring any packet loss and thus
_W_max_ is undefined. In this special case, CUBIC switches to _W_max_ is undefined. In this special case, CUBIC switches to
congestion avoidance and increases its congestion window size using congestion avoidance and increases its congestion window size using
Figure 1, where _t_ is the elapsed time since the beginning of the Figure 1, where _t_ is the elapsed time since the beginning of the
skipping to change at page 21, line 26 skipping to change at page 22, line 26
| 10000 | 83333.3 | 2.0e-10 | 1.0e-7 | 2.9e-8 | | 10000 | 83333.3 | 2.0e-10 | 1.0e-7 | 2.9e-8 |
+-------------------+-----------+---------+---------+---------+ +-------------------+-----------+---------+---------+---------+
Table 3: Required packet loss rate for Reno TCP, HSTCP, and Table 3: Required packet loss rate for Reno TCP, HSTCP, and
CUBIC to achieve a certain throughput CUBIC to achieve a certain throughput
Table 3 describes the required packet loss rate for Reno TCP, HSTCP, Table 3 describes the required packet loss rate for Reno TCP, HSTCP,
and CUBIC to achieve a certain throughput. We use 1500-byte packets and CUBIC to achieve a certain throughput. We use 1500-byte packets
and an _RTT_ of 0.1 seconds. and an _RTT_ of 0.1 seconds.
Our test results in [HKLRX06] indicate that CUBIC uses the spare Our test results in [HLRX07] indicate that CUBIC uses the spare
bandwidth left unused by existing Reno TCP flows in the same bandwidth left unused by existing Reno TCP flows in the same
bottleneck link without taking away much bandwidth from the existing bottleneck link without taking away much bandwidth from the existing
flows. flows.
5.3. Difficult Environments 5.3. Difficult Environments
CUBIC is designed to remedy the poor performance of Reno in fast and CUBIC is designed to remedy the poor performance of Reno in fast and
long-distance networks. long-distance networks.
5.4. Investigating a Range of Environments 5.4. Investigating a Range of Environments
There is decade-long deployment experience with CUBIC on the CUBIC has been extensively studied using simulations, testbed
Internet. CUBIC has also been extensively studied by using both NS-2 emulations, Internet experiments, and Internet measurements, covering
simulation and testbed experiments, covering a wide range of network a wide range of network environments
environments. More information can be found in [HKLRX06]. [HLRX07][H16][CEHRX09][HR11][BSCLU13][LBEWK16]. They have
convincingly demonstrated that CUBIC delivers substantial benefits
over classical Reno congestion control [RFC5681].
Same as Reno, CUBIC is a loss-based congestion control algorithm. Same as Reno, CUBIC is a loss-based congestion control algorithm.
Because CUBIC is designed to be more aggressive (due to a faster Because CUBIC is designed to be more aggressive (due to a faster
window increase function and bigger multiplicative decrease factor) window increase function and bigger multiplicative decrease factor)
than Reno in fast and long-distance networks, it can fill large drop- than Reno in fast and long-distance networks, it can fill large drop-
tail buffers more quickly than Reno and increases the risk of a tail buffers more quickly than Reno and increases the risk of a
standing queue [RFC8511]. In this case, proper queue sizing and standing queue [RFC8511]. In this case, proper queue sizing and
management [RFC7567] could be used to mitigate the risk to some management [RFC7567] could be used to mitigate the risk to some
extent and reduce the packet queuing delay. Also, in large-BDP extent and reduce the packet queuing delay. Also, in large-BDP
networks after a congestion event, CUBIC, due its cubic window networks after a congestion event, CUBIC, due its cubic window
skipping to change at page 22, line 33 skipping to change at page 23, line 35
exponentially increases the transmission timer for each packet exponentially increases the transmission timer for each packet
retransmission while congestion persists. retransmission while congestion persists.
5.6. Fairness within the Alternative Congestion Control Algorithm 5.6. Fairness within the Alternative Congestion Control Algorithm
CUBIC ensures convergence of competing CUBIC flows with the same RTT CUBIC ensures convergence of competing CUBIC flows with the same RTT
in the same bottleneck links to an equal throughput. When competing in the same bottleneck links to an equal throughput. When competing
flows have different RTT values, their throughput ratio is linearly flows have different RTT values, their throughput ratio is linearly
proportional to the inverse of their RTT ratios. This is true proportional to the inverse of their RTT ratios. This is true
independently of the level of statistical multiplexing on the link. independently of the level of statistical multiplexing on the link.
The convergence time depends on the network environments (e.g.,
bandwidth, RTT) and the level of statistical multiplexing, as
mentioned in Section 3.4.
5.7. Performance with Misbehaving Nodes and Outside Attackers 5.7. Performance with Misbehaving Nodes and Outside Attackers
This is not considered in the current CUBIC design. This is not considered in the current CUBIC design.
5.8. Behavior for Application-Limited Flows 5.8. Behavior for Application-Limited Flows
CUBIC does not increase its congestion window size if a flow is A flow is application-limited if it is currently sending less than
currently limited by the application instead of the congestion what is allowed by the congestion window. This can happen if the
window. Section 4.2 requires that _t_ in Figure 1 does not include flow is limited by either the sender application or the receiver
application-limited periods, such as idle periods, otherwise application (via the receiver advertised window) and thus sends less
W_cubic(_t_) might be very high after restarting from these periods. data than what is allowed by the sender's congestion window.
CUBIC does not increase its congestion window if a flow is
application-limited. Section 4.2 requires that _t_ in Figure 1 does
not include application-limited periods, such as idle periods,
otherwise W_cubic(_t_) might be very high after restarting from these
periods.
5.9. Responses to Sudden or Transient Events 5.9. Responses to Sudden or Transient Events
If there is a sudden increase in capacity, e.g., due to variable If there is a sudden increase in capacity, e.g., due to variable
radio capacity, a routing change, or a mobility event, CUBIC is radio capacity, a routing change, or a mobility event, CUBIC is
designed to utilize the newly available capacity faster than Reno. designed to utilize the newly available capacity faster than Reno.
On the other hand, if there is a sudden decrease in capacity, CUBIC On the other hand, if there is a sudden decrease in capacity, CUBIC
reduces more slowly than Reno. This remains true whether or not reduces more slowly than Reno. This remains true whether or not
CUBIC is in Reno-friendly mode and whether or not fast convergence is CUBIC is in Reno-friendly mode and whether or not fast convergence is
skipping to change at page 23, line 34 skipping to change at page 24, line 46
This document does not require any IANA actions. This document does not require any IANA actions.
8. References 8. References
8.1. Normative References 8.1. Normative References
[I-D.ietf-tcpm-hystartplusplus] [I-D.ietf-tcpm-hystartplusplus]
Balasubramanian, P., Huang, Y., and M. Olson, "HyStart++: Balasubramanian, P., Huang, Y., and M. Olson, "HyStart++:
Modified Slow Start for TCP", Work in Progress, Internet- Modified Slow Start for TCP", Work in Progress, Internet-
Draft, draft-ietf-tcpm-hystartplusplus-03, 25 July 2021, Draft, draft-ietf-tcpm-hystartplusplus-04, 23 January
<https://datatracker.ietf.org/doc/html/draft-ietf-tcpm- 2022, <https://datatracker.ietf.org/doc/html/draft-ietf-
hystartplusplus-03>. tcpm-hystartplusplus-04>.
[RFC2119] Bradner, S., "Key words for use in RFCs to Indicate [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate
Requirement Levels", BCP 14, RFC 2119, Requirement Levels", BCP 14, RFC 2119,
DOI 10.17487/RFC2119, March 1997, DOI 10.17487/RFC2119, March 1997,
<https://www.rfc-editor.org/rfc/rfc2119>. <https://www.rfc-editor.org/rfc/rfc2119>.
[RFC2883] Floyd, S., Mahdavi, J., Mathis, M., and M. Podolsky, "An
Extension to the Selective Acknowledgement (SACK) Option
for TCP", RFC 2883, DOI 10.17487/RFC2883, July 2000,
<https://www.rfc-editor.org/rfc/rfc2883>.
[RFC3168] Ramakrishnan, K., Floyd, S., and D. Black, "The Addition [RFC3168] Ramakrishnan, K., Floyd, S., and D. Black, "The Addition
of Explicit Congestion Notification (ECN) to IP", of Explicit Congestion Notification (ECN) to IP",
RFC 3168, DOI 10.17487/RFC3168, September 2001, RFC 3168, DOI 10.17487/RFC3168, September 2001,
<https://www.rfc-editor.org/rfc/rfc3168>. <https://www.rfc-editor.org/rfc/rfc3168>.
[RFC4015] Ludwig, R. and A. Gurtov, "The Eifel Response Algorithm
for TCP", RFC 4015, DOI 10.17487/RFC4015, February 2005,
<https://www.rfc-editor.org/rfc/rfc4015>.
[RFC5033] Floyd, S. and M. Allman, "Specifying New Congestion [RFC5033] Floyd, S. and M. Allman, "Specifying New Congestion
Control Algorithms", BCP 133, RFC 5033, Control Algorithms", BCP 133, RFC 5033,
DOI 10.17487/RFC5033, August 2007, DOI 10.17487/RFC5033, August 2007,
<https://www.rfc-editor.org/rfc/rfc5033>. <https://www.rfc-editor.org/rfc/rfc5033>.
[RFC5348] Floyd, S., Handley, M., Padhye, J., and J. Widmer, "TCP [RFC5348] Floyd, S., Handley, M., Padhye, J., and J. Widmer, "TCP
Friendly Rate Control (TFRC): Protocol Specification", Friendly Rate Control (TFRC): Protocol Specification",
RFC 5348, DOI 10.17487/RFC5348, September 2008, RFC 5348, DOI 10.17487/RFC5348, September 2008,
<https://www.rfc-editor.org/rfc/rfc5348>. <https://www.rfc-editor.org/rfc/rfc5348>.
skipping to change at page 25, line 7 skipping to change at page 26, line 31
RACK-TLP Loss Detection Algorithm for TCP", RFC 8985, RACK-TLP Loss Detection Algorithm for TCP", RFC 8985,
DOI 10.17487/RFC8985, February 2021, DOI 10.17487/RFC8985, February 2021,
<https://www.rfc-editor.org/rfc/rfc8985>. <https://www.rfc-editor.org/rfc/rfc8985>.
[RFC9002] Iyengar, J., Ed. and I. Swett, Ed., "QUIC Loss Detection [RFC9002] Iyengar, J., Ed. and I. Swett, Ed., "QUIC Loss Detection
and Congestion Control", RFC 9002, DOI 10.17487/RFC9002, and Congestion Control", RFC 9002, DOI 10.17487/RFC9002,
May 2021, <https://www.rfc-editor.org/rfc/rfc9002>. May 2021, <https://www.rfc-editor.org/rfc/rfc9002>.
8.2. Informative References 8.2. Informative References
[CEHRX07] Cai, H., Eun, D., Ha, S., Rhee, I., and L. Xu, "Stochastic [BSCLU13] Belhareth, S., Sassatelli, L., Collange, D., Lopez-
Ordering for Internet Congestion Control and its Pacheco, D., and G. Urvoy-Keller, "Understanding TCP cubic
Applications", IEEE INFOCOM 2007 - 26th IEEE International performance in the cloud: A mean-field approach", 2013
Conference on Computer Communications, IEEE 2nd International Conference on Cloud
DOI 10.1109/infcom.2007.111, 2007, Networking (CloudNet), DOI 10.1109/cloudnet.2013.6710576,
<https://doi.org/10.1109/infcom.2007.111>. November 2013,
<https://doi.org/10.1109/cloudnet.2013.6710576>.
[CEHRX09] Cai, H., Eun, D., Ha, S., Rhee, I., and L. Xu, "Stochastic
convex ordering for multiplicative decrease internet
congestion control", Computer Networks Vol. 53, pp.
365-381, DOI 10.1016/j.comnet.2008.10.012, February 2009,
<https://doi.org/10.1016/j.comnet.2008.10.012>.
[FHP00] Floyd, S., Handley, M., and J. Padhye, "A Comparison of [FHP00] Floyd, S., Handley, M., and J. Padhye, "A Comparison of
Equation-Based and AIMD Congestion Control", May 2000, Equation-Based and AIMD Congestion Control", May 2000,
<https://www.icir.org/tfrc/aimd.pdf>. <https://www.icir.org/tfrc/aimd.pdf>.
[GV02] Gorinsky, S. and H. Vin, "Extended Analysis of Binary [GV02] Gorinsky, S. and H. Vin, "Extended Analysis of Binary
Adjustment Algorithms", Technical Report TR2002-29, Adjustment Algorithms", Technical Report TR2002-29,
Department of Computer Sciences, The University of Department of Computer Sciences, The University of
Texas at Austin, 11 August 2002, Texas at Austin, 11 August 2002,
<https://www.cs.utexas.edu/ftp/techreports/tr02-39.ps.gz>. <https://www.cs.utexas.edu/ftp/techreports/tr02-39.ps.gz>.
[HKLRX06] Ha, S., Kim, Y., Le, L., Rhee, I., and L. Xu, "A Step [H16] Sangtae Ha, ., "Simulation, Testbed, and Deployment
toward Realistic Performance Evaluation of High-Speed TCP Testing Results of CUBIC", 3 November 2016,
Variants", International Workshop on Protocols for Fast <https://web.archive.org/web/20161118125842/
Long-Distance Networks, February 2006, http://netsrv.csc.ncsu.edu/wiki/index.php/TCP_Testing>.
<https://pfld.net/2006/paper/s2_03.pdf>.
[HR08] Ha, S. and I. Rhee, "Hybrid Slow Start for High-Bandwidth [HLRX07] Ha, S., Le, L., Rhee, I., and L. Xu, "Impact of background
and Long-Distance Networks", International Workshop traffic on performance of high-speed TCP variant
on Protocols for Fast Long-Distance Networks, March 2008, protocols", Computer Networks Vol. 51, pp. 1748-1762,
<http://www.hep.man.ac.uk/g/GDARN-IT/pfldnet2008/paper/ DOI 10.1016/j.comnet.2006.11.005, May 2007,
Sangate_Ha%20Final.pdf>. <https://doi.org/10.1016/j.comnet.2006.11.005>.
[HR11] Ha, S. and I. Rhee, "Taming the elephants: New TCP slow
start", Computer Networks Vol. 55, pp. 2092-2110,
DOI 10.1016/j.comnet.2011.01.014, June 2011,
<https://doi.org/10.1016/j.comnet.2011.01.014>.
[HRX08] Ha, S., Rhee, I., and L. Xu, "CUBIC: a new TCP-friendly [HRX08] Ha, S., Rhee, I., and L. Xu, "CUBIC: a new TCP-friendly
high-speed TCP variant", ACM SIGOPS Operating Systems high-speed TCP variant", ACM SIGOPS Operating Systems
Review Vol. 42, pp. 64-74, DOI 10.1145/1400097.1400105, Review Vol. 42, pp. 64-74, DOI 10.1145/1400097.1400105,
July 2008, <https://doi.org/10.1145/1400097.1400105>. July 2008, <https://doi.org/10.1145/1400097.1400105>.
[K03] Kelly, T., "Scalable TCP: improving performance in [K03] Kelly, T., "Scalable TCP: improving performance in
highspeed wide area networks", ACM SIGCOMM Computer highspeed wide area networks", ACM SIGCOMM Computer
Communication Review Vol. 33, pp. 83-91, Communication Review Vol. 33, pp. 83-91,
DOI 10.1145/956981.956989, April 2003, DOI 10.1145/956981.956989, April 2003,
<https://doi.org/10.1145/956981.956989>. <https://doi.org/10.1145/956981.956989>.
[LBEWK16] Lukaseder, T., Bradatsch, L., Erb, B., Van Der Heijden,
R., and F. Kargl, "A Comparison of TCP Congestion Control
Algorithms in 10G Networks", 2016 IEEE 41st Conference on
Local Computer Networks (LCN), DOI 10.1109/lcn.2016.121,
November 2016, <https://doi.org/10.1109/lcn.2016.121>.
[LIU16] Liu, K. and J. Lee, "On Improving TCP Performance over [LIU16] Liu, K. and J. Lee, "On Improving TCP Performance over
Mobile Data Networks", IEEE Transactions on Mobile Mobile Data Networks", IEEE Transactions on Mobile
Computing Vol. 15, pp. 2522-2536, Computing Vol. 15, pp. 2522-2536,
DOI 10.1109/tmc.2015.2500227, October 2016, DOI 10.1109/tmc.2015.2500227, October 2016,
<https://doi.org/10.1109/tmc.2015.2500227>. <https://doi.org/10.1109/tmc.2015.2500227>.
[RFC3465] Allman, M., "TCP Congestion Control with Appropriate Byte [RFC3465] Allman, M., "TCP Congestion Control with Appropriate Byte
Counting (ABC)", RFC 3465, DOI 10.17487/RFC3465, February Counting (ABC)", RFC 3465, DOI 10.17487/RFC3465, February
2003, <https://www.rfc-editor.org/rfc/rfc3465>. 2003, <https://www.rfc-editor.org/rfc/rfc3465>.
skipping to change at page 27, line 11 skipping to change at page 29, line 6
"TCP Alternative Backoff with ECN (ABE)", RFC 8511, "TCP Alternative Backoff with ECN (ABE)", RFC 8511,
DOI 10.17487/RFC8511, December 2018, DOI 10.17487/RFC8511, December 2018,
<https://www.rfc-editor.org/rfc/rfc8511>. <https://www.rfc-editor.org/rfc/rfc8511>.
[RFC9000] Iyengar, J., Ed. and M. Thomson, Ed., "QUIC: A UDP-Based [RFC9000] Iyengar, J., Ed. and M. Thomson, Ed., "QUIC: A UDP-Based
Multiplexed and Secure Transport", RFC 9000, Multiplexed and Secure Transport", RFC 9000,
DOI 10.17487/RFC9000, May 2021, DOI 10.17487/RFC9000, May 2021,
<https://www.rfc-editor.org/rfc/rfc9000>. <https://www.rfc-editor.org/rfc/rfc9000>.
[SXEZ19] Sun, W., Xu, L., Elbaum, S., and D. Zhao, "Model-Agnostic [SXEZ19] Sun, W., Xu, L., Elbaum, S., and D. Zhao, "Model-Agnostic
and Efficient Exploration of Numerical State Space of and Efficient Exploration of Numerical Congestion Control
Real-World TCP Congestion Control Implementations", USENIX State Space of Real-World TCP Implementations", IEEE/ACM
NSDI 2019, February 2019, Transactions on Networking Vol. 29, pp. 1990-2004,
<https://www.usenix.org/system/files/nsdi19-sun.pdf>. DOI 10.1109/tnet.2021.3078161, October 2021,
<https://doi.org/10.1109/tnet.2021.3078161>.
[XHR04] Xu, L., Harfoush, K., and I. Rhee, "Binary Increase [XHR04] Xu, L., Harfoush, K., and I. Rhee, "Binary increase
Congestion Control (BIC) for Fast Long-Distance Networks", congestion control (BIC) for fast long-distance networks",
IEEE INFOCOM 2004, DOI 10.1109/infcom.2004.1354672, March IEEE INFOCOM 2004, DOI 10.1109/infcom.2004.1354672, n.d.,
2004, <https://doi.org/10.1109/infcom.2004.1354672>. <https://doi.org/10.1109/infcom.2004.1354672>.
Appendix A. Acknowledgments Appendix A. Acknowledgments
Richard Scheffenegger and Alexander Zimmermann originally co-authored Richard Scheffenegger and Alexander Zimmermann originally co-authored
[RFC8312]. [RFC8312].
These individuals suggested improvements to this document: These individuals suggested improvements to this document:
* Bob Briscoe * Bob Briscoe
skipping to change at page 28, line 4 skipping to change at page 29, line 49
* Matt Olson * Matt Olson
* Michael Welzl * Michael Welzl
* Mirja Kuehlewind * Mirja Kuehlewind
* Mohit P. Tahiliani * Mohit P. Tahiliani
* Neal Cardwell * Neal Cardwell
* Praveen Balasubramanian * Praveen Balasubramanian
* Richard Scheffenegger * Richard Scheffenegger
* Rod Grimes * Rod Grimes
* Tom Henderson * Tom Henderson
* Tom Petch * Tom Petch
* Wesley Rosenblum * Wesley Rosenblum
* Yoshifumi Nishida * Yoshifumi Nishida
* Yuchung Cheng * Yuchung Cheng
Appendix B. Evolution of CUBIC Appendix B. Evolution of CUBIC
B.1. Since draft-ietf-tcpm-rfc8312bis-04 B.1. Since draft-ietf-tcpm-rfc8312bis-05
* Clarify meaning of "application-limited" in Section 5.8 (#137
(https://github.com/NTAP/rfc8312bis/issues/137)
* Brief discussion of convergence in Section 5.6 (#96
(https://github.com/NTAP/rfc8312bis/issues/96))
* Add more test results to Section 5 and update some references (#91
(https://github.com/NTAP/rfc8312bis/issues/91))
* Change wording around setting ssthresh (#131
(https://github.com/NTAP/rfc8312bis/issues/131))
B.2. Since draft-ietf-tcpm-rfc8312bis-04
* Fix incorrect math (#106 (https://github.com/NTAP/rfc8312bis/ * Fix incorrect math (#106 (https://github.com/NTAP/rfc8312bis/
issues/106)) issues/106))
* Update RFC5681 (#99 (https://github.com/NTAP/rfc8312bis/ * Update RFC5681 (#99 (https://github.com/NTAP/rfc8312bis/
issues/99)) issues/99))
* Rephrase text around algorithmic alternatives, add HyStart++ (#85 * Rephrase text around algorithmic alternatives, add HyStart++ (#85
(https://github.com/NTAP/rfc8312bis/issues/85), #86 (https://github.com/NTAP/rfc8312bis/issues/85), #86
(https://github.com/NTAP/rfc8312bis/issues/86), #90 (https://github.com/NTAP/rfc8312bis/issues/86), #90
skipping to change at page 29, line 15 skipping to change at page 31, line 27
* Clarify text around queuing and slow adaptation of CUBIC in * Clarify text around queuing and slow adaptation of CUBIC in
wireless environments (#94 (https://github.com/NTAP/rfc8312bis/ wireless environments (#94 (https://github.com/NTAP/rfc8312bis/
issues/94)) issues/94))
* Set lower bound of cwnd to 1 MSS and use retransmit timer * Set lower bound of cwnd to 1 MSS and use retransmit timer
thereafter (#83 (https://github.com/NTAP/rfc8312bis/issues/83)) thereafter (#83 (https://github.com/NTAP/rfc8312bis/issues/83))
* Use FlightSize instead of cwnd to update ssthresh (#114 * Use FlightSize instead of cwnd to update ssthresh (#114
(https://github.com/NTAP/rfc8312bis/issues/114)) (https://github.com/NTAP/rfc8312bis/issues/114))
B.2. Since draft-ietf-tcpm-rfc8312bis-03 * Create new subsections for spurious timeouts and spurious loss via
ACK (#90 (https://github.com/NTAP/rfc8312bis/issues/90))
B.3. Since draft-ietf-tcpm-rfc8312bis-03
* Remove reference from abstract (#82 * Remove reference from abstract (#82
(https://github.com/NTAP/rfc8312bis/pull/82)) (https://github.com/NTAP/rfc8312bis/pull/82))
B.3. Since draft-ietf-tcpm-rfc8312bis-02 B.4. Since draft-ietf-tcpm-rfc8312bis-02
* Description of packet loss rate _p_ (#65 * Description of packet loss rate _p_ (#65
(https://github.com/NTAP/rfc8312bis/issues/65)) (https://github.com/NTAP/rfc8312bis/issues/65))
* Clarification of TCP Friendly Equation for ABC and Delayed ACK * Clarification of TCP Friendly Equation for ABC and Delayed ACK
(#66 (https://github.com/NTAP/rfc8312bis/issues/66)) (#66 (https://github.com/NTAP/rfc8312bis/issues/66))
* add applicability to QUIC and SCTP (#61 * add applicability to QUIC and SCTP (#61
(https://github.com/NTAP/rfc8312bis/issues/61)) (https://github.com/NTAP/rfc8312bis/issues/61))
skipping to change at page 29, line 46 skipping to change at page 32, line 14
* clarify _cwnd_ growth in convex region (#69 * clarify _cwnd_ growth in convex region (#69
(https://github.com/NTAP/rfc8312bis/issues/69)) (https://github.com/NTAP/rfc8312bis/issues/69))
* add guidance for using bytes and mention that segments count is * add guidance for using bytes and mention that segments count is
decimal (#67 (https://github.com/NTAP/rfc8312bis/issues/67)) decimal (#67 (https://github.com/NTAP/rfc8312bis/issues/67))
* add loss events detected by RACK and QUIC loss detection (#62 * add loss events detected by RACK and QUIC loss detection (#62
(https://github.com/NTAP/rfc8312bis/issues/62)) (https://github.com/NTAP/rfc8312bis/issues/62))
B.4. Since draft-ietf-tcpm-rfc8312bis-01 B.5. Since draft-ietf-tcpm-rfc8312bis-01
* address Michael Scharf's editorial suggestions. (#59 * address Michael Scharf's editorial suggestions. (#59
(https://github.com/NTAP/rfc8312bis/issues/59)) (https://github.com/NTAP/rfc8312bis/issues/59))
* add "Note to the RFC Editor" about removing underscores * add "Note to the RFC Editor" about removing underscores
B.5. Since draft-ietf-tcpm-rfc8312bis-00 B.6. Since draft-ietf-tcpm-rfc8312bis-00
* use updated xml2rfc with better text rendering of subscripts * use updated xml2rfc with better text rendering of subscripts
B.6. Since draft-eggert-tcpm-rfc8312bis-03 B.7. Since draft-eggert-tcpm-rfc8312bis-03
* fix spelling nits * fix spelling nits
* rename to draft-ietf * rename to draft-ietf
* define _W_max_ more clearly * define _W_max_ more clearly
B.7. Since draft-eggert-tcpm-rfc8312bis-02 B.8. Since draft-eggert-tcpm-rfc8312bis-02
* add definition for segments_acked and alpha__aimd_. (#47 * add definition for segments_acked and alpha__aimd_. (#47
(https://github.com/NTAP/rfc8312bis/issues/47)) (https://github.com/NTAP/rfc8312bis/issues/47))
* fix a mistake in _W_max_ calculation in the fast convergence * fix a mistake in _W_max_ calculation in the fast convergence
section. (#51 (https://github.com/NTAP/rfc8312bis/issues/51)) section. (#51 (https://github.com/NTAP/rfc8312bis/issues/51))
* clarity on setting _ssthresh_ and _cwnd_start_ during * clarity on setting _ssthresh_ and _cwnd_start_ during
multiplicative decrease. (#53 (https://github.com/NTAP/rfc8312bis/ multiplicative decrease. (#53 (https://github.com/NTAP/rfc8312bis/
issues/53)) issues/53))
B.8. Since draft-eggert-tcpm-rfc8312bis-01 B.9. Since draft-eggert-tcpm-rfc8312bis-01
* rename TCP-Friendly to AIMD-Friendly and rename Standard TCP to * rename TCP-Friendly to AIMD-Friendly and rename Standard TCP to
AIMD TCP to avoid confusion as CUBIC has been widely used on the AIMD TCP to avoid confusion as CUBIC has been widely used on the
Internet. (#38 (https://github.com/NTAP/rfc8312bis/issues/38)) Internet. (#38 (https://github.com/NTAP/rfc8312bis/issues/38))
* change introductory text to reflect the significant broader * change introductory text to reflect the significant broader
deployment of CUBIC on the Internet. (#39 deployment of CUBIC on the Internet. (#39
(https://github.com/NTAP/rfc8312bis/issues/39)) (https://github.com/NTAP/rfc8312bis/issues/39))
* rephrase introduction to avoid referring to variables that have * rephrase introduction to avoid referring to variables that have
not been defined yet. not been defined yet.
B.9. Since draft-eggert-tcpm-rfc8312bis-00 B.10. Since draft-eggert-tcpm-rfc8312bis-00
* acknowledge former co-authors (#15 * acknowledge former co-authors (#15
(https://github.com/NTAP/rfc8312bis/issues/15)) (https://github.com/NTAP/rfc8312bis/issues/15))
* prevent _cwnd_ from becoming less than two (#7 * prevent _cwnd_ from becoming less than two (#7
(https://github.com/NTAP/rfc8312bis/issues/7)) (https://github.com/NTAP/rfc8312bis/issues/7))
* add list of variables and constants (#5 * add list of variables and constants (#5
(https://github.com/NTAP/rfc8312bis/issues/5), #6 (https://github.com/NTAP/rfc8312bis/issues/5), #6
(https://github.com/NTAP/rfc8312bis/issues/6)) (https://github.com/NTAP/rfc8312bis/issues/6))
skipping to change at page 31, line 28 skipping to change at page 33, line 47
* note for Fast Recovery during _cwnd_ decrease due to congestion * note for Fast Recovery during _cwnd_ decrease due to congestion
event (#11 (https://github.com/NTAP/rfc8312bis/issues/11)) event (#11 (https://github.com/NTAP/rfc8312bis/issues/11))
* add section for spurious congestion events (#23 * add section for spurious congestion events (#23
(https://github.com/NTAP/rfc8312bis/issues/23)) (https://github.com/NTAP/rfc8312bis/issues/23))
* initialize _W_est_ after timeout and remove variable * initialize _W_est_ after timeout and remove variable
_W_(last_max)_ (#28 (https://github.com/NTAP/rfc8312bis/ _W_(last_max)_ (#28 (https://github.com/NTAP/rfc8312bis/
issues/28)) issues/28))
B.10. Since RFC8312 B.11. Since RFC8312
* converted to Markdown and xml2rfc v3 * converted to Markdown and xml2rfc v3
* updated references (as part of the conversion) * updated references (as part of the conversion)
* updated author information * updated author information
* various formatting changes * various formatting changes
* move to Standards Track * move to Standards Track
B.11. Since the Original Paper B.12. Since the Original Paper
CUBIC has gone through a few changes since the initial release CUBIC has gone through a few changes since the initial release
[HRX08] of its algorithm and implementation. Below we highlight the [HRX08] of its algorithm and implementation. Below we highlight the
differences between its original paper and [RFC8312]. differences between its original paper and [RFC8312].
* The original paper [HRX08] includes the pseudocode of CUBIC * The original paper [HRX08] includes the pseudocode of CUBIC
implementation using Linux's pluggable congestion control implementation using Linux's pluggable congestion control
framework, which excludes system-specific optimizations. The framework, which excludes system-specific optimizations. The
simplified pseudocode might be a good source to start with and simplified pseudocode might be a good source to start with and
understand CUBIC. understand CUBIC.
 End of changes. 54 change blocks. 
127 lines changed or deleted 217 lines changed or added

This html diff was produced by rfcdiff 1.48. The latest version is available from http://tools.ietf.org/tools/rfcdiff/