Re: [tsvwg] L4S vs SCE

[Roland Bless <[e-mail]>]

Hi,

On 20.11.19 at 13:40 De Schepper, Koen (Nokia - BE/Antwerp) wrote:
>
> I agree with Ingemar. Related to the chances for failing: as history
> learned us, the only real world chance for failing would be that the
> technology would not be taken up in deployments. This whole codepoint
> conflict is mainly damaging/creating this deployment 'issue'. The
> other so called show stoppers are less of an issue for most people
> that worked on CCs. Blowing 'issues' (not yet
> implemented/integrated/tested code and bugs) out of proportion is not
> the way to go forward here.
>
>  
>
> To give an example:
>
> As mentioned in another thread already: the L4S codepoint, where
> there's a new queue and new CCs, creates a clean-slate opportunity.
> This is an opportunity to select new desired behavior.
>
Yes, but as I also expressed my concerns w.r.t. the L4S codepoint
earlier, at the cost of binding this to a quite fixed set of L4S
behaviors and "burning" the last ECT codepoint. Personally, I like
concepts with a little bit more potential to be useful for future
development (evolvability) of congestion controls, e.g., BBRv2 and LoLa
could also benefit from an SCE-like marking...

Therefore, a compromise could be to define ECT(1) with "some congestion
experienced" semantics that could also be used for L4S.
In order to further disambiguate and use the particular L4S semantics of
Dual Queue AQM etc. one could use an additional DSCP as
L4S classifier: I currently don't see the need to follow the argument of
providing L4S orthogonal to DiffServ (and for me L4S is specifying a PHB).
In case L4S doesn't take off one can at least use the SCE semantics and
reassign the DSCP (or in case of success make it later orthogonal
to DiffServ). Due to the newly opened pool of DSCPs one may have a
chance of finding a DSCP that has a high probability of
getting through from end-to-end.

> One of these opportunities here is to make L4S_TCP less RTT dependent.
> There have been many working implementations for less RTT-dependent
> CCs in the past. One that is widely deployed is Cubic, which is doing
> this for getting more throughput for longer RTTs. The only reason why
> it didn’t fly for lower RTTs on the current Internet is that they
> would hurt themselves (get lower throughput, competing with Reno).
>
>  
>
> As we are able to define a new independent behavior and the RTT
> dependence in L4S is taken serious (some even call it a show stopper)
> this is even a must do opportunity for L4S. It is all a matter of
> perspective: show-stopper <-> opportunity.
>
Achieving RTT fairness isn't easy, but the Fair Flow Balancing mechanism
of TCP LoLa is one
example of how one can achieve this...

 
>
> Let it be clear, I am not against adopting new ideas for
> implementation of congestions controls and AQMs, as long as they are
> not contesting with the code point and requirements that were selected
> and agreed on as part of the L4S BoF. I'm also not against the recent
> vibrant energy, interest and engagement of people, but I think we can
> better use this energy in making things work. As you noticed, we can
> use the help to speed things up on the open source part.
>
I didn't see the adoption of L4S work at the BOF as an agreement to
accept everything as proposed rather than
an agreement to discuss and work on the issues, and to come up with a
potentially different solution as it has been
at the time of the BOF.

Regards,
 Roland

>  
>
> Regards,
>
> Koen.
>
>  
>
> *From:* Ingemar Johansson S <[e-mail]>
> *Sent:* Wednesday, November 20, 2019 11:50 AM
> *To:* Kyle Rose <[e-mail]>; Ingemar Johansson S
> <[e-mail]>
> *Cc:* Sebastian Moeller <[e-mail]>; G Fairhurst
> <[e-mail]>; [e-mail]; [e-mail]; De
> Schepper, Koen (Nokia - BE/Antwerp)
> <[e-mail]>; [e-mail];
> Ingemar Johansson S <[e-mail]>
> *Subject:* RE: [tsvwg] L4S vs SCE
>
>  
>
> Hi
>
>  
>
> So given the imagined outcome that L4S fails.. two scenarios
>
>  
>
> If other SDOs or developers don’t pick up L4S then things are quite
> simple I guess, just declare the L4S drafts as deprecated, or is there
> more to do ?
>
>  
>
> But, if e.g. 3GPP somehow thinks that this is a good idea and adopts
> it.. Will the IETF send a message (LS?) to 3GPP with the message
> “please stop using L4S”, is this even a reasonable scenario?. After
> all, the fact that it is picked up by other SDOs, speaks against a
> failure ?
>
>  
>
> /Ingemar
>
>  
>
> *From:* Kyle Rose <[e-mail] <mailto:[e-mail]>>
> *Sent:* den 20 november 2019 11:14
> *To:* Ingemar Johansson S
> <ingemar.s.johansson=[e-mail]
> <mailto:ingemar.s.johansson=[e-mail]>>
> *Cc:* Sebastian Moeller <[e-mail] <mailto:[e-mail]>>; G
> Fairhurst <[e-mail] <mailto:[e-mail]>>;
> [e-mail] <mailto:[e-mail]>; Ingemar Johansson S
> <[e-mail]
> <mailto:[e-mail]>>; [e-mail]
> <mailto:[e-mail]>; De Schepper, Koen (Koen)
> <[e-mail] <mailto:[e-mail]>>;
> [e-mail] <mailto:[e-mail]>
> *Subject:* Re: [tsvwg] L4S vs SCE
>
>  
>
> On Wed, Nov 20, 2019 at 6:04 PM Ingemar Johansson S
> <ingemar.s.johansson=[e-mail]
> <mailto:[e-mail]>> wrote:
>
>     >       How do you expect an industry/commercial roll-out of L4S
>     > technology to behave, if the L4S experiment should terminated
>     without
>     > adoption as a standard track RFC? Are they supposed to phase-out
>     using
>     > ECT(1) as well, or is it understood that deployed L4S instances
>     continue using
>     > L4S methods?
>
>     [IJ] The premise would be that L4S is declared a failure. I doubt
>     that anybody has a good enough crystal ball to predict what
>     happens. First it is necessary to come to the conclusion that L4S
>     is a failure and I would argue that we are not yet there and I
>     don't currently see that coming. Before that possible event I
>     don't see it meaningful to speculate.
>
>  
>
> I'm going to have to go ahead and disagree with you strongly here.
> Given the potential consequences of cleaning up after a failed
> experiment without a plan worked out beforehand, this blithe approach
> is simply not acceptable.
>
>  
>
> In lots of cases, experiments are easy to terminate in an obvious way:
> for example, in one typical case, a code point can simply be
> abandoned, or (even better) a pollutable experimental code point
> returned to the available pool after some time. If that were the case
> here, it would not be difficult to enumerate a sequence of steps
> required to do so. It doesn't appear that's the case, however, so all
> the more reason to make sure we address this as part of the experiment
> setup.
>
>  
>
> A launch escape system of the necessary complexity should be a
> requirement of any experimental deployment. In this case, that might
> circumscribe the scope of the experiment itself.
>
>  
>
> Kyle
>
>  
>