[Detnet] Re: WG adoption poll: joung-detnet-stateless-fair-queuing-05

[Jinoo Joung <[e-mail]>]

Hello Chairs, Scott, and WG,

Please note that draft-joung-detnet-stateless-fair-queuing is revised and
uploaded as 07.
In line with Scott's suggestions, the new version includes:
1) Abbreviations
2) New references including RFC8200, MPLS-related IETF drafts, and two
ITU-T Recommendations.
3) Minimized 6.1. Framework and other content that are duplicative with
ITU-T Recommendations. Only those that are essential to understand the
whole picture are left.
4) Elaborated 6.3.1 Metadata; 6.3.2. Header format; 6.3.3. Admission
control; 7. Approximate C-SCORE

In 6.3.1 and 6.3.2, the methods for carrying metadata with IPv6 and MPLS
MNA Sub-stack are described in detail.
6.3.3 delineates two distinct admission control methodologies, providing
for the selection of either procedure based on specific implementation
requirements.
Section 7. Approximate C-SCORE specifies the nodal architecture and
algorithms to approximate C-SCORE with rotating strict priority schedulers.
The E2E latency bound, in a complete math expression, is also given.
This approximation serves to alleviate the sorting bottleneck intrinsic to
C-SCORE,
bypassing the intensive priority queue operations while maintaining
near-optimal alignment with theoretical Finish Time objectives.

Best regards,
Jinoo (on behalf of authors)

====================================

Internet-Draft draft-joung-detnet-stateless-fair-queuing-07.txt is now
available. It is a work item of the Deterministic Networking (DETNET) WG of
the IETF.

   Title:   Latency Guarantee with Stateless Fair Queuing
   Authors: Jinoo Joung
            Jeong-dong Ryoo
            Taesik Cheung
            Yizhou Li
            Peng Liu
   Name:    draft-joung-detnet-stateless-fair-queuing-07.txt
   Pages:   23
   Dates:   2026-02-20

Abstract:

   This document specifies the framework and the operational procedure
   for deterministic networking with a set of rate based work conserving
   packet schedulers.  The framework guarantees end-to-end (E2E) latency
   bounds to flows.  The schedulers in core nodes do not need to
   maintain flow states.  Instead, the entrance node of a flow marks an
   ideal service completion time according to a fluid model, called
   Finish Time (FT), of a packet in the packet header.  The subsequent
   core nodes update the FT by adding the delay factor, which is a
   function of the flow and the nodes.  The packets in the queue of the
   scheduler are served in the ascending order of FT.  This mechanism is
   called the stateless fair queuing.  The result is that flows are
   isolated from each other almost perfectly.  The latency bound of a
   flow depends only on the flow's intrinsic parameters such as the
   maximum burst size and the service rate, except the link capacities
   and the maximum packet length among other flows sharing each output
   link with the flow.  Furthermore, this document specifies an
   approximation of stateless fair queuing implemented via a strict
   priority (SP) scheduler.  This approach maintains a guaranteed end-
   to-end (E2E) latency bound.

The IETF datatracker status page for this Internet-Draft is:
https://datatracker.ietf.org/doc/draft-joung-detnet-stateless-fair-queuing/

There is also an HTMLized version available at:
https://datatracker.ietf.org/doc/html/draft-joung-detnet-stateless-fair-queuing-07

A diff from the previous version is available at:
https://author-tools.ietf.org/iddiff?url2=draft-joung-detnet-stateless-fair-queuing-07

Internet-Drafts are also available by rsync at:
rsync.ietf.org::internet-drafts


On Fri, Jan 23, 2026 at 10:52 PM Janos Farkas <[e-mail]>
wrote:

> Hi,
>
>
>
> Scott,
>
> Thank you very mush for your guidelines! Much appreciated.
>
>
>
> Jinoo,
>
> Great to see agreement.
>
> Could you please revise the draft along the guidelines provided? Once
> updated per the discussion below, we can publish it as a WG document.
>
>
>
> Thank you,
>
> Lou and Janos
>
>
>
> *From:* Jinoo Joung <[e-mail]>
> *Sent:* Tuesday, January 20, 2026 4:23 PM
> *To:* Scott Mansfield <scott.mansfield=[e-mail]>
> *Cc:* [e-mail]
> *Subject:* [Detnet] Re: WG adoption poll:
> joung-detnet-stateless-fair-queuing-05
>
>
>
> Hello Scott,
>
> Thanks for your valuable suggestions.
> I deeply appreciate your efforts to resolve this issue in a fair and
> reasonable manner.
> I agree with your proposed action.
>
> I have answered your guidelines in-line marked with JJ. Please take a look.
>
> In short, there is a clear gap between the ITU-T work and the IETF
> document.
> While ITU-T work focuses on the requirements and framework, the IETF
> document will elaborate on IP/MPLS specific information for deployment,
> including header fields and formats, protocols for admission control and
> time-difference compensation.
> It will also include implementation details for approximating fair queuing
> with a strict priority scheduler.
>
> Many thanks and best regards,
> Jinoo
>
>
>
> On Tue, Jan 20, 2026 at 6:11 AM Scott Mansfield <scott.mansfield=
> [e-mail]> wrote:
>
> Hi,
>
>
>
> Speaking as a liaison manager between the IETF and ITU-T, it is important
> to note a few guidelines.  Following this discussion, I observe there is
> concern that there is work being duplicated.  In order to make all parties
> comfortable I would suggest the following:
>
>
>
> Potential Action:
>
> The process of “gap analysis” is used a lot in the ITU-T to ensure new
> work items do not duplicate work already done by the industry recognized
> center of competence.  The reverse is true as well.  I would want to see
> the gap that the IETF is filling here and how it augments or modifies what
> the ITU-T has already done.  If there are modifications to the ITU-T work
> needed, I would like to see the authors plans to progress that work in the
> ITU-T.
>
>
>
> JJ: There is a clear gap, as I stated in the main text of this mail. The
> IETF draft will augment the ITU-T work. There is no need to modify the
> ITU-T work. It serves as a solid foundation for the IETF draft. In my
> opinion, the implementation specific issues in the IETF draft are not
> appropriate for standardization in ITU-T.
>
>
>
> Guidelines To Note:
> - It is appropriate for the IETF to take ITU-T work and augment it as long
> as the augmentation is in the scope and mandate of IETF.  Also, the work
> should not fundamentally change (or be backwards incompatible).  It is
> imperative that the IETF document be clear what ITU-T work it is using and
> properly reference the work.
>
>
>
> JJ: I believe the IP/MPLS issues, protocols, and implementation
> specifications are well within the scope of IETF. While the work proceeds
> in IETF, no changes will be introduced. The ITU-T works will be referenced
> as well.
>
>
>
> - If there is anything being added in the IETF that requires a fundamental
> change to the ITU-T work, that work needs to be liaised/coordinated with
> the ITU-T.  If what the IETF is doing is explaining how to use the ITU-T
> work in the context of IETF (profiles etc.) that is ok.
>
>
>
> JJ: Again, the ITU-T work won't be changed. The IETF draft will explain
> how to use the fundamentals described in the ITU-T work.
>
>
>
> - If the IETF work is really changing the math, that should be coordinated
> too.  If the ITU-T work missed something, or is not appropriate for the
> IETF use-case, again, coordination is needed.  Since (at least) some of the
> authors are the same in both groups, this is the best way to gain consensus
> by ensuring the process of each group is followed.
>
>
>
> JJ: I definitely agree with you.
>
>
>
> I find coordination and cooperation is best handled when people work
> together and have people that attend each organization and follow the
> processes of the organization when progressing the documents.  Liaison can
> be used, but they tend to be slow.  The IETF DetNet mailing list is the
> proper place to have this discussion especially if there are participants
> in this working group that need clarity to ensure work is not duplicated.
>
>
>
> Many thanks,
>
> -scott.
>
>
>
>
>
>
>
> [Detnet] Re: WG adoption poll: joung-detnet-stateless-fair-queuing-05
>
> Jinoo Joung <[e-mail]> Fri, 02 January 2026 22:16 UTCShow header
> <https://mailarchive.ietf.org/arch/msg/detnet/Ea4920SPZmgxU-0npKrss00EI7Y/>
>
> Hello Mike, it's great to hear from you.
>
> Thanks for the valuable feedback.
>
> It seems that your major concern is the discrepancy between the standards.
>
>
>
> But I assure you that there is no conflict.
>
> Because ITU-T did not allow math equations in the Requirements section,
>
> I had to divide one equation into two or three sentences.
>
> That is why it seems the number of requirements are different.
>
> If you find any discrepancies, just let me know.
>
>
>
> Y.3148 also deals with considerations for metadata, but it still is at an
>
> academic level.
>
> These two ITU-T standards are strictly based on the paper:
>
> https://ieeexplore.ieee.org/document/10261190
>
>
>
> Best regards,
>
> Jinoo
>
>
>
>
>
>
>
>
>
> On Sat, Jan 3, 2026 at 4:09 AM Mike Blanche <[e-mail]> wrote:
>
>
>
> >
>
> > Hi Jinoo and all,
>
> >
>
> > Although the C-SCORE approach may be an appropriate solution within the
>
> > agreed taxonomy, I am still concerned about duplicating work already done
>
> > elsewhere.
>
> >
>
> > What happens if the IETF draft diverges from the ITU Recommendation?
>
> > Y.3129 makes 9 "requirements", whereas the current draft has 5
> requirements
>
> > and 3 conditions. There is already one discrepancy between the two
>
> > documents in the description of the maths. This may lead to inconsistent
> or
>
> > incompatible implementations depending on which document implementers
>
> > follow.
>
> >
>
> > I see the new draft  (-06) focuses more on the necessary metadata. This
> is
>
> > also already documented in ITU-T Y.3148, which covers metadata creation
> and
>
> > updating procedures in depth. Again, how will it be ensured that this
> draft
>
> > does not diverge from or conflict with the existing ITU Recommendation?
>
> >
>
> > The latest draft still does not cite either ITU Recommendation Y.3129 or
>
> > Y.3148.
>
> >
>
> > Thanks,
>
> >
>
> > Mike
>
> >
>
> > On Tue, 23 Dec 2025 at 00:08, Jinoo Joung <[e-mail]> wrote:
>
> >
>
> >>
>
> >> Hello Mike, thanks for the comment regarding the work in ITU-T.
>
> >> I understand your concern.
>
> >>
>
> >> Hello WG and Mike,
>
> >>
>
> >> The C-SCORE draft, as indicated in the version 6, will have the
> following
>
> >> details in the future, once accepted as the WG draft.
>
> >> 1) MPLS and IP specific protocol details: This includes the header
>
> >> fields/formats, interactions between nodes and between control/data
> planes.
>
> >> 2) Implementation issues regarding the priority queue, which is
> currently
>
> >> required to sort the packets in FT orders: This includes an
> approximation
>
> >> technique based on strict priority schedulers.
>
> >>
>
> >> Y.3129 is focused on the requirements and provides theoretical
>
> >> justifications but the C-SCORE draft targets on the solution.
>
> >> Some framework texts are covered in both. But they are necessary
>
> >> information for readers in order to understand the design principles of
> the
>
> >> solution in the C-SCORE draft.
>
> >>
>
> >> C-SCORE is not a new invention.
>
> >> It is an outcome of the numerous researches conducted during late '90 ~
>
> >> early '00s, which was the peak period of Internet research.
>
> >> There are more than hundreds of papers that are related to the fair
>
> >> queuing, after all.
>
> >>
>
> >> I believe it is worth being standardized by such virtues:
>
> >> 1) flow isolation capability, which protects from other flows'
>
> >> join/leave, bursts, or even malicious behaviours
>
> >> 2) scalability due to its stateless nature
>
> >> 3) flexibility from network asymmetry such as link capacity variations
>
> >> and uneven propagation delays
>
> >> 4) simple admission criteria, which allows incremental flow acceptance,
>
> >> without modifying existing network setup
>
> >> 5) small average delay due to work conserving nature,
>
> >> just to mention a few. :-)
>
> >>
>
> >> Best regards,
>
> >> Jinoo (on behalf of authors)
>
> >>
>
> >>
>
> >>
>
> >> On Fri, Dec 12, 2025 at 12:10 AM Mike Blanche <[e-mail]>
>
> >> wrote:
>
> >>
>
> >>>
>
> >>> Hi all,
>
> >>>
>
> >>> This proposal appears very similar to ITU-T Recommendation Y.3129,
> which
>
> >>> I believe was developed by some of the same authors.
>
> >>>
>
> >>> https://www.itu.int/rec/T-REC-Y.3129-202404-I/en
>
> >>>
>
> >>> The abstract text of the I-D is virtually identical to the summary of
>
> >>> Y.3129. In addition, significant parts of the content are very similar
> -
>
> >>> e.g. parts of section 6.3 of the draft are word-for-word the same as
> parts
>
> >>> of section 8 of the ITU-T Recommendation.
>
> >>>
>
> >>> The ITU-T Recommendation is missing from the draft's References
> section.
>
> >>>
>
> >>> I am unclear why we are considering duplicating work that appears to be
>
> >>> already completed at the ITU.
>
> >>>
>
> >>> Best regards,
>
> >>>
>
> >>> Mike
>
> >>>
>
> >>> On Fri, 5 Dec 2025 at 15:49, Janos Farkas <Janos.Farkas=
>
> >>> [e-mail]> wrote:
>
> >>>
>
> >>>> Hi,
>
> >>>>
>
> >>>>
>
> >>>>
>
> >>>> This email begins a 4-week adoption poll for:
>
> >>>>
>
> >>>>
>
> >>>>
>
> >>>>
>
> >>>>
> https://datatracker.ietf.org/doc/draft-joung-detnet-stateless-fair-queuing/05/
>
> >>>>
>
> >>>>
>
> >>>>
>
> >>>> No IPR has been disclosed for this document.
>
> >>>>
>
> >>>>
>
> >>>>
>
> >>>> Please voice your support or technical objections to adoption on the
>
> >>>>
>
> >>>> list by the end of the day (any time zone) January 2nd.
>
> >>>>
>
> >>>>
>
> >>>>
>
> >>>> As a reminder this document is part of the larger set of adoption
> calls
>
> >>>>
>
> >>>> of the documents discussed at IETF 124:
>
> >>>>
>
> >>>>
>
> >>>>
>
> >>>>
>
> >>>>
> https://datatracker.ietf.org/doc/draft-joung-detnet-stateless-fair-queuing/05
>
> >>>>
>
> >>>>
>
> >>>>
> https://datatracker.ietf.org/doc/draft-peng-detnet-deadline-based-forwarding/18
>
> >>>>
>
> >>>>
>
> >>>>
> https://datatracker.ietf.org/doc/draft-peng-detnet-packet-timeslot-mechanism/13
>
> >>>>
>
> >>>> https://datatracker.ietf.org/doc/draft-eckert-detnet-tcqf/09
>
> >>>>
>
> >>>> https://datatracker.ietf.org/doc/draft-eckert-detnet-glbf/06
>
> >>>>
>
> >>>>
> https://datatracker.ietf.org/doc/draft-ryoo-detnet-ontime-forwarding/04
>
> >>>>
>
> >>>> https://datatracker.ietf.org/doc/draft-ryoo-detnet-nscore/02
>
> >>>>
>
> >>>>
>
> >>>>
>
> >>>> Thank you,
>
> >>>>
>
> >>>> János (as Co-chair)
>
> >>>>
>
> >>>>
>
> >>>> _______________________________________________
>
> >>>> detnet mailing list -- [e-mail]
>
> >>>> To unsubscribe send an email to [e-mail]
>
> >>>>
>
> >>> _______________________________________________
>
> >>> detnet mailing list -- [e-mail]
>
> >>> To unsubscribe send an email to [e-mail]
>
> >>>
>
> >>
>
>    - [Detnet] WG adoption poll: joung-detnet-stateless…
>    <https://mailarchive.ietf.org/arch/msg/detnet/Q2JXJlfyvd-WzGkkqtRfCr3yHOs/>  Janos
>    Farkas
>    - [Detnet] Re: WG adoption poll: joung-detnet-state…
>    <https://mailarchive.ietf.org/arch/msg/detnet/tqDrY-nmIDalPRGf5OMD0_xWvOU/>  Mike
>    Blanche
>    - [Detnet] Re: WG adoption poll: joung-detnet-state…
>    <https://mailarchive.ietf.org/arch/msg/detnet/S51eXeGt_UcT_1cVqUZrUxejR3U/>
>      song.xueyan2
>    - [Detnet] Re: WG adoption poll: joung-detnet-state…
>    <https://mailarchive.ietf.org/arch/msg/detnet/3MdTx2InCj1rbv31dZrqBFzfio8/>  Jinoo
>    Joung
>    - [Detnet] Re: WG adoption poll: joung-detnet-state…
>    <https://mailarchive.ietf.org/arch/msg/detnet/BE3XHigD2aMc8-NgWPZhBWRG_ic/>  Yizhou
>    Li
>    - [Detnet] Re: WG adoption poll: joung-detnet-state…
>    <https://mailarchive.ietf.org/arch/msg/detnet/l6I5fRbyBp1QQZEYv-4sSuSve_E/>
>      유연철
>    - [Detnet] Re: WG adoption poll: joung-detnet-state…
>    <https://mailarchive.ietf.org/arch/msg/detnet/7ksa7vAMmQ3sehku5DO8LfONpPg/>  Jeong-dong
>    Ryoo
>    - [Detnet] Re: WG adoption poll: joung-detnet-state…
>    <https://mailarchive.ietf.org/arch/msg/detnet/2GqbxqQFYw2wYOv1c8UnF3alGns/>
>      xiong.quan
>    - [Detnet] Re: WG adoption poll: joung-detnet-state…
>    <https://mailarchive.ietf.org/arch/msg/detnet/_G3i5tAVEVx2ptaZ5waatbMJo6Y/>
>      정태식
>    - [Detnet] Re: WG adoption poll: joung-detnet-state…
>    <https://mailarchive.ietf.org/arch/msg/detnet/FFbeeIUbtYnhRQ04YDfa13_3hrc/>  Mike
>    Blanche
>    - [Detnet] Re: WG adoption poll: joung-detnet-state…
>    <https://mailarchive.ietf.org/arch/msg/detnet/Ea4920SPZmgxU-0npKrss00EI7Y/>  Jinoo
>    Joung
>    - [Detnet] Re: WG adoption poll: joung-detnet-state…
>    <https://mailarchive.ietf.org/arch/msg/detnet/24tmXTeBKDecp-cQi4rhK1H0jGk/>
>      [e-mail]
>    - [Detnet] Re: WG adoption poll: joung-detnet-state…
>    <https://mailarchive.ietf.org/arch/msg/detnet/AhQOorY02WR6zJowkcLrXGPqk-E/>  Xuesong
>    Geng
>
>
>
> _______________________________________________
> detnet mailing list -- [e-mail]
> To unsubscribe send an email to [e-mail]
>
>