Re: [tsvwg] Comments on draft-ietf-tsvwg-ecn-l4s-id-05

[Ingemar Johansson S <[e-mail]>]

Hi Bob



Please see inline, marked [IJ2]



/Ingemar



From: Bob Briscoe <[e-mail]>
Sent: den 15 november 2018 16:04
To: Ingemar Johansson S <[e-mail]>
Cc: [e-mail]; Magnus Westerlund <[e-mail]>; 
'[e-mail]' <[e-mail]>
Subject: Re: Comments on draft-ietf-tsvwg-ecn-l4s-id-05



Ingemar,

On 11/11/2018 20:28, Ingemar Johansson S wrote:

Q3. When you say the following:

RFC8298 (SCReAM) currently only specifies classic ECN handling, but running 
code of SCReAM found at https://github.com/EricssonResearch/scream has a 
working L4S implementation.

are you saying RFC8298 only specifies classic ECN feedback? Or are you now 
talking about the L4S sender setting ECT(1) and using scalable (L4S) 
congestion control? Or both?

[IJ2] RFC8298 does only specify classic ECN back-off 
https://tools.ietf.org/html/rfc8298#page-17 . The RFC8298 does not mention L4S 
at all and therefore it does not distinguish between ECT(0) and ECT(1).

[IJ] RFC8298 does only specify the behavior assuming classic ECN feedback 
(ECT(0) set). The running code has an additional L4S mode that implements a 
scalable reaction to congestion. It is possible that this can added to an 
updated RFC8298 when/if it is updated from experimental to standards track.


If you are solely talking about the sender behaviour here, I would rather move 
this to where we talk about setting the codepoint or the congestion control.


[BB] Your answer still seems to mix feedback (receiver and sender) with 
congestion response (sender).

[IJ2] Yes, possibly , feedback is not specified in RFC8298, there is however a 
list of requirements https://tools.ietf.org/html/rfc8298#page-28



Here's my question written more clearly:

If both ends already support avtcore-cc-feedback-message, can one end 
unilaterally implement the L4S congestion response (and therefore set ECT(1) 
on outgoing packets)?

[IJ2] Yes, that is possible as avtcore-cc-feedback-message  indicates the ECN 
bits for each received packet.


Or did you have to enhance the avtcore-cc-feedback-message code (for writing 
feedback at one end and reading it at the other), before one end (or both) 
could add the L4S congestion response?

[IJ2] No



The stated goal of avtcore-cc-feedback-message is:

   To enable algorithm evolution as well as
   interoperability across designs (e.g., different rate adaptation
   algorithms), it is highly desirable to have generic congestion
   control feedback format.

If you had to enhance the feedback format for L4S, that would imply the 
feedback message is still not generic enough.

[IJ2] No, enhancement of the avtcore-cc-feedback-message  is not necessary, 
the only thing that is required is that the sender is aware that it sets 
ECT(1) and applies a scalable congestion control.




Bob




-- 
________________________________________________________________
Bob Briscoe                               http://bobbriscoe.net/