Re: [tsvwg] New Version Notification for draft-herbert-tcp-in-udp-00.txt

[Michael Welzl <[e-mail]>]

Dear Tom, dear everyone else @ TSVWG,

I’m (somewhat obviously, if you see what I write below) strongly in favour of this draft - and also happy to help with it.  This idea (in some form) has been proposed multiple times over the years, and never took off; perhaps the time is now right?
Below is some material.


Earlier similar proposals:
====================

The first appearance I know of:  https://datatracker.ietf.org/doc/html/draft-denis-udp-transport-00 <https://datatracker.ietf.org/doc/html/draft-denis-udp-transport-00>   (likely it isn’t the first to exist!)
Much later:  https://datatracker.ietf.org/doc/html/draft-cheshire-tcp-over-udp-00 <https://datatracker.ietf.org/doc/html/draft-cheshire-tcp-over-udp-00>
and finally, our own effort:  https://datatracker.ietf.org/doc/html/draft-welzl-tcp-ccc <https://datatracker.ietf.org/doc/html/draft-welzl-tcp-ccc>
which I presented in ICCRG, at IETF-95:  https://folk.universitetetioslo.no/michawe/research/publications/ietf95-iccrg-tcp-in-udp.pdf <https://folk.universitetetioslo.no/michawe/research/publications/ietf95-iccrg-tcp-in-udp.pdf>



Work showing benefits:
===================

There may be several reasons to do this. Our own interest, back then, was in implementing (single-path!) coupled congestion control in the style of RFC 8699 for TCP.
Such coupling can only work when we know that the connections traverse the same bottleneck - either via measurement (RFC 8382), or because, as with this encapsulation, they carry the same outer 5-tuple.
Our coupling is lightweight, as it doesn’t require re-vamping the entire congestion control code like the congestion manager. It still yields all the benefits of a single congestion control instance, just like a multi-streaming transport would have: lower latency and packet loss from less competition, and the ability to execute precise prioritization.

This was done in the context of Safiqul Islam’s Ph.D. thesis. These publications describe the benefits:

Safiqul Islam, Michael Welzl, Kristian Hiorth, David Hayes, Grenville Armitage, Stein Gjessing: "ctrlTCP: Reducing Latency through Coupled, Heterogeneous Multi-Flow TCP Congestion Control", IEEE INFOCOM Global Internet Symposium (GI) workshop (GI 2018), Honolulu, HI, April 2018. DOI 10.1109/INFCOMW.2018.8406887
https://ieeexplore.ieee.org/document/8406887 <https://ieeexplore.ieee.org/document/8406887>
Preprint: https://folk.universitetetioslo.no/michawe/research/publications/ctrltcp-infocom-gi-2018.pdf <https://folk.universitetetioslo.no/michawe/research/publications/ctrltcp-infocom-gi-2018.pdf>

and, focusing on the benefit of ramping up cwnd faster when joining other connections:
Safiqul Islam, Michael Welzl: "Start Me Up: Determining and Sharing TCP's Initial Congestion Window", ACM, IRTF, ISOC Applied Networking Research Workshop 2016 (ANRW 2016), Berlin, Germany, 16 July 2016. DOI https://doi.org/10.1145/2959424.2959440 <https://doi.org/10.1145/2959424.2959440>
https://dl.acm.org/authorize?N16076 <https://dl.acm.org/authorize?N16076>


Code:
======

We have ( ...somewhere. I’ll be happy to dig for it if someone wants it! ) a FreeBSD implementation of both the congestion control coupling and the encapsulation, and it’s described in more detail here:
https://www.duo.uio.no/handle/10852/51926 <https://www.duo.uio.no/handle/10852/51926>


Cheers,
Michael



> On 12 Aug 2023, at 00:22, Tom Herbert <tom=[e-mail]> wrote:
> 
> Hi,
> 
> This is a draft describing how TCP could be encapsulated in UDP. This
> would be useful if UDP really is to be the "new network layer" and we
> want to use UDP Options with TCP to carry network layer options.
> 
> Thanks,
> Tom
> 
> 
> ---------- Forwarded message ---------
> From: <[e-mail]>
> Date: Fri, Aug 11, 2023 at 3:19 PM
> Subject: New Version Notification for draft-herbert-tcp-in-udp-00.txt
> To: Tom Herbert <[e-mail]>
> 
> 
> 
> A new version of I-D, draft-herbert-tcp-in-udp-00.txt
> has been successfully submitted by Tom Herbert and posted to the
> IETF repository.
> 
> Name:           draft-herbert-tcp-in-udp
> Revision:       00
> Title:          TCP-in-UDP Encapsulation
> Document date:  2023-08-11
> Group:          Individual Submission
> Pages:          10
> URL:            https://www.ietf.org/archive/id/draft-herbert-tcp-in-udp-00.txt
> Status:         https://datatracker.ietf.org/doc/draft-herbert-tcp-in-udp/
> Htmlized:       https://datatracker.ietf.org/doc/html/draft-herbert-tcp-in-udp
> 
> 
> Abstract:
>   This document specifies a method of encapsulating TCP protocol
>   packets within UDP headers.  TCP-in-UDP is useful in situations where
>   network capabilities specific to UDP can be leveraged for TCP
>   packets.
> 
> 
> 
> 
> The IETF Secretariat
>