CUBIC for Fast and Long-Distance Networks
draft-ietf-tcpm-rfc8312bis-15

[Ballot comment]
After reading the ballots, I went to go see what all the fuss is about with the Shepherd Writeup -- and fully support everyone's comments; it was nicely detailed and helpful.

I'd also like to thank Bo Wu for the OpsDir writeup, and also, of course, the authors for writing this. I didn't understand much of the deep details / math, but it sounded plausible and used many big words, so... ¯\_(ツ)_/¯

[Ballot comment]
Thanks for the solid work on this document and also wanted to say, thanks for the great shepherd writeup that was detailed clear and concise.

[Ballot comment]
It's nice to see such expressive math in an RFC (even if I don't comprehend most of it, I'm sure the target audience appreciates it too)

[Ballot comment]
Thank you for the work on this document.

Many thanks to Spencer Dawkins for his ART ART review: https://mailarchive.ietf.org/arch/msg/art/_BMAGlC1TXKYcLy0HC1yF6ZR11M/ , and to the author for addressing Spencer's comments.

[Ballot comment]
# Internet AD comments for draft-ietf-tcpm-rfc8312bis-14
CC @ekline

## Nits

### S4.4, S4.5

* The text file version seems to have rendered the "cwnd" in the demoninator
  as "culvert", which was pretty confusing.  :-)

### S5.1

* "networks with a short RTTs" -> "networks with short RTTs", probably

Request for Last Call review by ARTART Completed: Ready with Nits. Reviewer: Spencer Dawkins. Sent review to list. Submission of review completed at an earlier date.

(Via drafts-lastcall@iana.org): IESG/Authors/WG Chairs:

The IANA Functions Operator has reviewed draft-ietf-tcpm-rfc8312bis-14, which is currently in Last Call, and has the following comments:

We understand that this document doesn't require any registry actions.

While it's often helpful for a document's IANA Considerations section to remain in place upon publication even if there are no actions, if the authors strongly prefer to remove it, we do not object.

If this assessment is not accurate, please respond as soon as possible.

For definitions of IANA review states, please see:

https://datatracker.ietf.org/help/state/draft/iana-review

Thank you,

Sabrina Tanamal
Lead IANA Services Specialist

The following Last Call announcement was sent out (ends 2022-12-19):

From: The IESG
To: IETF-Announce
CC: draft-ietf-tcpm-rfc8312bis@ietf.org, martin.h.duke@gmail.com, nsd.ietf@gmail.com, tcpm-chairs@ietf.org, tcpm@ietf.org
Reply-To: last-call@ietf.org
Sender:
Subject: Last Call:  (CUBIC for Fast and Long-Distance Networks) to Proposed Standard


The IESG has received a request from the TCP Maintenance and Minor Extensions
WG (tcpm) to consider the following document: - 'CUBIC for Fast and
Long-Distance Networks'
  as Proposed Standard

The IESG plans to make a decision in the next few weeks, and solicits final
comments on this action. Please send substantive comments to the
last-call@ietf.org mailing lists by 2022-12-19. Exceptionally, comments may
be sent to iesg@ietf.org instead. In either case, please retain the beginning
of the Subject line to allow automated sorting.

Abstract


  CUBIC is a standard TCP congestion control algorithm that uses a
  cubic function instead of a linear congestion window increase
  function to improve scalability and stability over fast and long-
  distance networks.  CUBIC has been adopted as the default TCP
  congestion control algorithm by the Linux, Windows, and Apple stacks.

  This document updates the specification of CUBIC to include
  algorithmic improvements based on these implementations and recent
  academic work.  Based on the extensive deployment experience with
  CUBIC, it also moves the specification to the Standards Track,
  obsoleting RFC 8312.  This also requires updating RFC 5681, to allow
  for CUBIC's occasionally more aggressive sending behavior.




The file can be obtained via
https://datatracker.ietf.org/doc/draft-ietf-tcpm-rfc8312bis/



No IPR declarations have been submitted directly on this I-D.

Document Shepherd Write-Up for Group Documents
This version is dated 4 July 2022.

# Document History

1) Does the working group (WG) consensus represent the strong concurrence of a
few individuals, with others being silent, or did it reach broad agreement?

There was a broad consensus in the WG on the document including the implementers
from major OSes and QUIC community.


2) Was there controversy about particular points, or were there decisions where
the consensus was particularly rough?

Although there was a broad consensus in the WG, it was not unanimous.
This draft specifies CUBIC congestion control algorithm as an updated version of RFC8312,
which will promote the specification to a proposed standard. This means we will have
the second standard congestion control algorithm for TCP in addition to Reno which was
published 20 years ago. Because of the importance of the document, there were arguments
on the points listed below.

1: RFC5033 provides a guideline and a process for considering new congestion control
  algorithms within the IETF. One argument was that the draft has not been
  through the entire process described in the RFC.

2: CUBIC utilizes Reno-friendly model for controlling transfer rate in order to
  behave mostly fairly when it competes with Reno in low BDP environments.
  One argument was that the paper which originally proposed the model has not provided
  convincing explanations that the model can be well-suited for Today's Internet.
  Also, a recent test result suggests that CUBIC looks fair to Reno in those low
  BDP environments.

3: CUBIC uses 0.7 as multiplicative decrease factor for congestion window while Reno
  uses 0.5. In low BDP environments, Reno-friendly model is used to compensate the
  aggressiveness by adopting slower congestion window growth rate than Reno. However,
  one argument was the high multiplicative decrease factor will still be
  aggressive in some situations, which leads to unfair resource sharing with Reno.

As a result of discussions, we've reached a rough consensus to publish the document
and decided to continue the discussions on these points for the future version of
the documents. The summarized rationales behind the decision are the followings
while more detailed explanations were added in the document.
     
* CUBIC already has been deployed globally for decades and we have not observed
  any evidences for the issues or potential risks for it in the past.

* We have a solid consensus that there is fairly low risk that any parts of
  CUBIC's logics lead to congestion collapse on the Internet.

* As CUBIC has been widely deployed as default congestion control scheme, Reno
  is becoming less used on the Internet. This would mean maintaining fairness with
  Reno becomes less important.

* It is not obvious that CUBIC go beyond the general congestion control principles
  outlined in RFC2914, so the process described in RFC5033 may not need to be applied.
  Also, the document provides explanations of the safety features in CUBIC to meet
  the guidelines in RFC5033.


3) Has anyone threatened an appeal or otherwise indicated extreme discontent? If
so, please summarize the areas of conflict in separate email messages to the
responsible Area Director. (It should be in a separate email because this
questionnaire is publicly available.)

No one has threatened an appeal. We haven't seen any other conflicts or
discontents except the points described in 2).


4) For protocol documents, are there existing implementations of the contents of
the document? Have a significant number of potential implementers indicated
plans to implement? Are any existing implementations reported somewhere,
either in the document itself (as RFC 7942 recommends) or elsewhere
(where)?

CUBIC has been one of the most deployed congestion control algorithm for TCP
which all major OSes have supported. Also, various QUIC implementations
support it as well.

# Additional Reviews

5) Do the contents of this document closely interact with technologies in other
IETF working groups or external organizations, and would it therefore benefit
from their review? Have those reviews occurred? If yes, describe which
reviews took place.

As CUBIC algorithm can be used in other transport protocols, We've received
various individual feedback from the experts of QUIC. Also, linux kernel
experts had provided feedback from the implementors' points of view.

6) Describe how the document meets any required formal expert review criteria,
such as the MIB Doctor, YANG Doctor, media type, and URI type reviews.

This document does not require any formal reviews.

7) If the document contains a YANG module, has the final version of the module
been checked with any of the recommended validation tools for syntax and
formatting validation? If there are any resulting errors or warnings, what is
the justification for not fixing them at this time? Does the YANG module
comply with the Network Management Datastore Architecture (NMDA) as specified
in RFC 8342?

The document does not contain a YANG module.


8) Describe reviews and automated checks performed to validate sections of the
final version of the document written in a formal language, such as XML code,
BNF rules, MIB definitions, CBOR's CDDL, etc.

No formal language verifications are necessary for the documents.


# Document Shepherd Checks

9) Based on the shepherd's review of the document, is it their opinion that this
document is needed, clearly written, complete, correctly designed, and ready
to be handed off to the responsible Area Director?

Yes, the document is clearly written and ready to be handed off.


10) Several IETF Areas have assembled lists of common issues that their
reviewers encounter. For which areas have such issues been identified
and addressed? For which does this still need to happen in subsequent
reviews?

No issue has been identified.

11) What type of RFC publication is being requested on the IETF stream (Best
Current Practice, Proposed Standard, Internet Standard,
Informational, Experimental or Historic)? Why is this the proper type
of RFC? Do all Datatracker state attributes correctly reflect this intent?

The intended status of the document is Proposed Standard as there is a strong consensus
in the WG. Datatracker state attributes correctly reflect it.

12) Have reasonable efforts been made to remind all authors of the intellectual
property rights (IPR) disclosure obligations described in BCP 79? To
the best of your knowledge, have all required disclosures been filed? If
not, explain why. If yes, summarize any relevant discussion, including links
to publicly-available messages when applicable.

We've confirmed from all authors that there is no IPR related to the contents
of the document.


13) Has each author, editor, and contributor shown their willingness to be
listed as such? If the total number of authors and editors on the front page
is greater than five, please provide a justification.

Yes, all five authors are willing to be listed as such.

14) Document any remaining I-D nits in this document. Simply running the idnits
tool is not enough; please review the "Content Guidelines" on
authors.ietf.org. (Also note that the current idnits tool generates
some incorrect warnings; a rewrite is underway.)

idnits 2.17.1 reports no errors in the document. There are some warnings, but
we've confirmed that we don't need to address them.


15) Should any informative references be normative or vice-versa? See the IESG
Statement on Normative and Informative References.

No, they are classified correctly.

16) List any normative references that are not freely available to anyone. Did
the community have sufficient access to review any such normative
references?

None

17) Are there any normative downward references (see RFC 3967 and BCP
97) that are not already listed in the DOWNREF registry? If so,
list them.

None

18) Are there normative references to documents that are not ready to be
submitted to the IESG for publication or are otherwise in an unclear state?
If so, what is the plan for their completion?

No such reference.

19) Will publication of this document change the status of any existing RFCs?

The document will obsolete RFC8312 as the new specification for CUBIC and
will move the specification to the Standards Track. It will also update some
restrictions specified in RFC5681 based on the differences between Reno and CUBIC.

20) Describe the document shepherd's review of the IANA considerations section,
especially with regard to its consistency with the body of the document.
Confirm that all aspects of the document requiring IANA assignments are
associated with the appropriate reservations in IANA registries. Confirm
that any referenced IANA registries have been clearly identified. Confirm
that each newly created IANA registry specifies its initial contents,
allocations procedures, and a reasonable name (see RFC 8126).

This document does not require any IANA actions.

21) List any new IANA registries that require Designated Expert Review for
future allocations. Are the instructions to the Designated Expert clear?
Please include suggestions of designated experts, if appropriate.

This document does not have any new IANA registries.

Document Shepherd Write-Up for Group Documents
This version is dated 4 July 2022.

# Document History

1) Does the working group (WG) consensus represent the strong concurrence of a
few individuals, with others being silent, or did it reach broad agreement?

There was a broad consensus in the WG on the document including the implementers
from major OSes and QUIC community.


2) Was there controversy about particular points, or were there decisions where
the consensus was particularly rough?

Although there was a broad consensus in the WG, it was not unanimous.
This draft specifies CUBIC congestion control algorithm as an updated version of RFC8312,
which will promote the specification to a proposed standard. This means we will have
the second standard congestion control algorithm for TCP in addition to Reno which was
published 20 years ago. Because of the importance of the document, there were arguments
on the points listed below.

1: RFC5033 provides a guideline and a process for considering new congestion control
  algorithms within the IETF. One argument was that the draft has not been
  through the entire process described in the RFC.

2: CUBIC utilizes Reno-friendly model for controlling transfer rate in order to
  behave mostly fairly when it competes with Reno in low BDP environments.
  One argument was that the paper which originally proposed the model has not provided
  convincing explanations that the model can be well-suited for Today's Internet.
  Also, a recent test result suggests that CUBIC looks fair to Reno in those low
  BDP environments.

3: CUBIC uses 0.7 as multiplicative decrease factor for congestion window while Reno
  uses 0.5. In low BDP environments, Reno-friendly model is used to compensate the
  aggressiveness by adopting slower congestion window growth rate than Reno. However,
  one argument was the high multiplicative decrease factor will still be
  aggressive in some situations, which leads to unfair resource sharing with Reno.

As a result of discussions, we've reached a rough consensus to publish the document
and decided to continue the discussions on these points for the future version of
the documents. The summarized rationales behind the decision are the followings
while more detailed explanations were added in the document.
     
* CUBIC already has been deployed globally for decades and we have not observed
  any evidences for the issues or potential risks for it in the past.

* We have a solid consensus that there is fairly low risk that any parts of
  CUBIC's logics lead to congestion collapse on the Internet.

* As CUBIC has been widely deployed as default congestion control scheme, Reno
  is becoming less used on the Internet. This would mean maintaining fairness with
  Reno becomes less important.

* It is not obvious that CUBIC go beyond the general congestion control principles
  outlined in RFC2914, so the process described in RFC5033 may not need to be applied.
  Also, the document provides explanations of the safety features in CUBIC to meet
  the guidelines in RFC5033.


3) Has anyone threatened an appeal or otherwise indicated extreme discontent? If
so, please summarize the areas of conflict in separate email messages to the
responsible Area Director. (It should be in a separate email because this
questionnaire is publicly available.)

No one has threatened an appeal. We haven't seen any other conflicts or
discontents except the points described in 2).


4) For protocol documents, are there existing implementations of the contents of
the document? Have a significant number of potential implementers indicated
plans to implement? Are any existing implementations reported somewhere,
either in the document itself (as RFC 7942 recommends) or elsewhere
(where)?

CUBIC has been one of the most deployed congestion control algorithm for TCP
which all major OSes have supported. Also, various QUIC implementations
support it as well.

# Additional Reviews

5) Do the contents of this document closely interact with technologies in other
IETF working groups or external organizations, and would it therefore benefit
from their review? Have those reviews occurred? If yes, describe which
reviews took place.

As CUBIC algorithm can be used in other transport protocols, We've received
various individual feedback from the experts of QUIC. Also, linux kernel
experts had provided feedback from the implementors' points of view.

6) Describe how the document meets any required formal expert review criteria,
such as the MIB Doctor, YANG Doctor, media type, and URI type reviews.

This document does not require any formal reviews.

7) If the document contains a YANG module, has the final version of the module
been checked with any of the recommended validation tools for syntax and
formatting validation? If there are any resulting errors or warnings, what is
the justification for not fixing them at this time? Does the YANG module
comply with the Network Management Datastore Architecture (NMDA) as specified
in RFC 8342?

The document does not contain a YANG module.


8) Describe reviews and automated checks performed to validate sections of the
final version of the document written in a formal language, such as XML code,
BNF rules, MIB definitions, CBOR's CDDL, etc.

No formal language verifications are necessary for the documents.


# Document Shepherd Checks

9) Based on the shepherd's review of the document, is it their opinion that this
document is needed, clearly written, complete, correctly designed, and ready
to be handed off to the responsible Area Director?

Yes, the document is clearly written and ready to be handed off.


10) Several IETF Areas have assembled lists of common issues that their
reviewers encounter. For which areas have such issues been identified
and addressed? For which does this still need to happen in subsequent
reviews?

No issue has been identified.

11) What type of RFC publication is being requested on the IETF stream (Best
Current Practice, Proposed Standard, Internet Standard,
Informational, Experimental or Historic)? Why is this the proper type
of RFC? Do all Datatracker state attributes correctly reflect this intent?

The intended status of the document is Proposed Standard as there is a strong consensus
in the WG. Datatracker state attributes correctly reflect it.

12) Have reasonable efforts been made to remind all authors of the intellectual
property rights (IPR) disclosure obligations described in BCP 79? To
the best of your knowledge, have all required disclosures been filed? If
not, explain why. If yes, summarize any relevant discussion, including links
to publicly-available messages when applicable.

We've confirmed from all authors that there is no IPR related to the contents
of the document.


13) Has each author, editor, and contributor shown their willingness to be
listed as such? If the total number of authors and editors on the front page
is greater than five, please provide a justification.

Yes, all five authors are willing to be listed as such.

14) Document any remaining I-D nits in this document. Simply running the idnits
tool is not enough; please review the "Content Guidelines" on
authors.ietf.org. (Also note that the current idnits tool generates
some incorrect warnings; a rewrite is underway.)

idnits 2.17.1 reports no errors in the document. There are some warnings, but
we've confirmed that we don't need to address them.


15) Should any informative references be normative or vice-versa? See the IESG
Statement on Normative and Informative References.

No, they are classified correctly.

16) List any normative references that are not freely available to anyone. Did
the community have sufficient access to review any such normative
references?

None

17) Are there any normative downward references (see RFC 3967 and BCP
97) that are not already listed in the DOWNREF registry? If so,
list them.

None

18) Are there normative references to documents that are not ready to be
submitted to the IESG for publication or are otherwise in an unclear state?
If so, what is the plan for their completion?

No such reference.

19) Will publication of this document change the status of any existing RFCs?

The document will obsolete RFC8312 as the new specification for CUBIC and
will move the specification to the Standards Track. It will also update some
restrictions specified in RFC5681 based on the differences between Reno and CUBIC.

20) Describe the document shepherd's review of the IANA considerations section,
especially with regard to its consistency with the body of the document.
Confirm that all aspects of the document requiring IANA assignments are
associated with the appropriate reservations in IANA registries. Confirm
that any referenced IANA registries have been clearly identified. Confirm
that each newly created IANA registry specifies its initial contents,
allocations procedures, and a reasonable name (see RFC 8126).

This document does not require any IANA actions.

21) List any new IANA registries that require Designated Expert Review for
future allocations. Are the instructions to the Designated Expert clear?
Please include suggestions of designated experts, if appropriate.

This document does not have any new IANA registries.