Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[alex.b...@ealdwulf.org.uk]

Hi Bob,


I note that all the non-dependent claims of US20170019343A1 (claims 1,14,22) 
seem to assume use of the proportional-integral controller (Note, I am not a 
lawyer, and especially not a patent lawyer). In Appendix B of 
draft-briscoe-tsvwg-aqm-dualq-coupled, an alternate algorithm 'Curvy RED' seems 
to replace PI, but it is noted that 'the Curvy RED algorithm has not been 
maintained to the same degree as the DualPI2 algorithm '.

Can you comment on whether the Curvy RED algorithm could form a 
non-patent-encumbered dualq? In particular:
 - Why wasn't curvy red further developed? Was it found to contain some 
deficiency? Are you intending to present it as an alternative?
 - Does Curvy RED actually completely replace PI?
 - Can we have reasonable assurance that no patents will surface covering Curvy 
RED?

Thanks,
Alex


On Wednesday, March 20, 2019, 11:29:38 PM GMT, Bob Briscoe 
 wrote: 






1/ In 2016, I arranged for the hire of a patent attorney to undertake the 
unusual step of filing a third party observation with the European Patent 
Office. This went through Al-Lu's patent application claim by claim pointing to 
prior art and giving the patent examiner all the arguments to reject each 
claim. However, the examiner chose to take very little note of it, which was 
disappointing and costly for us. The main prior art is:
    Gibbens, R.J. & Kelly, F.P., "On Packet Marking at Priority Queues," IEEE 
Transactions on Automatic Control 47(6):1016--1020 (June 2002)
The guys named as inventors in AL-Lu's filing published a paper on PI2 with me, 
in which we included a citation of this Gibbens paper as inspiration for the 
coupling. The Gibbens paper was already cited as background by other patents, 
so the EPO has it in their citation index.

The coupling was also based on my prior research with Mirja before I started 
working with the guys from Al-Lu in the RITE European Collaborative project. we 
had to go through a few rejections, but Mirja and I finally got this work 
published in 2014  - still before the priority date of the Al-Lu patent 
application:
    Kühlewind, M., Wagner, D.P., Espinosa, J.M.R. & Briscoe, B., "Using Data 
Center TCP (DCTCP) in the Internet," In: Proc. Third IEEE Globecom Workshop on 
Telecommunications Standards: From Research to Standards pp.583-588 (December 
2014)

2/ The only claim that I could not find prior art for (in the original EU 
filing) was a very specific claim about using a square root for the coupling. 
The Linux implementation runs this the other way round so that it only has to 
do a squaring. So I figured we were safe from that.

However, until just now, I had not noticed that Al-Lu has retrospectively 
re-written the claims in the US patent and in the EU patent application to 
claim this the other way round - as a squaring. And to claim the two random 
number trick. Both restructuring to use a squaring and the two random number 
trick were definitely my ideas (while working for BT in a collaboration with 
Al-Lu). I have emails to prove this (from memory they were actually both in the 
same email). This is important, because a patent has to be about mechanism, not 
algorithm.

3/ This is a positive development. It means this patent is on very shaky legal 
ground. I have been trying to put pressure on Nokia to license this royalty 
free. But now I see what they have done, I am going to have to get a different 
type of legal advice. 


___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Alex Burr]

Bob,
 Thanks, this is interesting. As you probably know this patent has come to
the attention of the Linux community and caused some concern:
https://lwn.net/Articles/784125/
so it's useful to know of a potential workaround.

Alex

On Mon, Mar 25, 2019 at 1:34 AM Bob Briscoe  wrote:

> Alex, inline...
>
> On 24/03/2019 21:15, alex.b...@ealdwulf.org.uk wrote:
> >
> > Hi Bob,
> >
> >
> > I note that all the non-dependent claims of US20170019343A1 (claims
> 1,14,22) seem to assume use of the proportional-integral controller (Note,
> I am not a lawyer, and especially not a patent lawyer).
> Yes, as I understand it, Nokia's intention with this filing was to cover
> use of the PI controller in particular, in combination with various
> other ideas.
>
> > In Appendix B of draft-briscoe-tsvwg-aqm-dualq-coupled, an alternate
> algorithm 'Curvy RED' seems to replace PI, but it is noted that 'the Curvy
> RED algorithm has not been maintained to the same degree as the DualPI2
> algorithm '.
> >
> > Can you comment on whether the Curvy RED algorithm could form a
> non-patent-encumbered dualq? In particular:
> >   - Why wasn't curvy red further developed? Was it found to contain some
> deficiency? Are you intending to present it as an alternative?
> We just didn't develop it further, cos we were getting better results
> with PI2. However, I am aware of a hardware implementation based on
> Curvy RED going on at the moment, and you will see there have recently
> been review comments on that Curvy RED appendix on the list.
>
> So, even tho PI might be better, Curvy RED (or another AQM) might be
> preferable for other reasons that performance (e.g. ease of
> implementation, or similarity to an existing hardware implementation).
>
> And indeed, there's nothing to stop anyone using other AQMs, either to
> work round the IPR, or because they're preferable in their own right -
> the DualQ Coupled AQM is intentionally a framework into which you drop 2
> AQMs.
>
> >   - Does Curvy RED actually completely replace PI?
> Yes.
> >   - Can we have reasonable assurance that no patents will surface
> covering Curvy RED?
> Well, I developed the idea of Curvy RED and I / my employer (BT) did not
> file any IPR at the time. I got approval to publish a tech report
> jointly with Al-Lu. http://bobbriscoe.net/pubs.html#CRED-insights
>
> That was May 2015, so given nothing has surfaced by now, there can't be
> anything from that time from us (where us = Al-Lu & BT).
>
> Of course, I cannot guarantee that there is not another patent in the
> system from some other random company that my searches haven't found.
> There are large numbers of AQM patents. Also, I cannot guarantee that an
> implementer working now isn't filing patents around their
> implementation. All we can do is publish as much as possible as early as
> possible to try to keep some areas of the field open.
>
>
> Bob
> >
> > Thanks,
> > Alex
> >
> >
> > On Wednesday, March 20, 2019, 11:29:38 PM GMT, Bob Briscoe <
> i...@bobbriscoe.net> wrote:
> >
> >
> >
> >
> >
> >
> > 1/ In 2016, I arranged for the hire of a patent attorney to undertake
> the unusual step of filing a third party observation with the European
> Patent Office. This went through Al-Lu's patent application claim by claim
> pointing to prior art and giving the patent examiner all the arguments to
> reject each claim. However, the examiner chose to take very little note of
> it, which was disappointing and costly for us. The main prior art is:
> >  Gibbens, R.J. & Kelly, F.P., "On Packet Marking at Priority
> Queues," IEEE Transactions on Automatic Control 47(6):1016--1020 (June 2002)
> > The guys named as inventors in AL-Lu's filing published a paper on PI2
> with me, in which we included a citation of this Gibbens paper as
> inspiration for the coupling. The Gibbens paper was already cited as
> background by other patents, so the EPO has it in their citation index.
> >
> > The coupling was also based on my prior research with Mirja before I
> started working with the guys from Al-Lu in the RITE European Collaborative
> project. we had to go through a few rejections, but Mirja and I finally got
> this work published in 2014  - still before the priority date of the Al-Lu
> patent application:
> >  Kühlewind, M., Wagner, D.P., Espinosa, J.M.R. & Briscoe, B., "Using
> Data Center TCP (DCTCP) in the Internet," In: Proc. Third IEEE Globecom
> Workshop on Telecommunications Standards: From Research to Standards
> pp.583-588 (December 2014)
> >
> > 2/ The only claim that I could not find prior art for (in the original
> EU filing) was a very specific claim about using a square root for the
> coupling. The Linux implementation runs this the other way round so that it
> only has to do a squaring. So I figured we were safe from that.
> >
> > However, until just now, I had not noticed that Al-Lu has
> retrospectively re-written the claims in the US patent and in the EU patent
> application to claim this the other way rou

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Rodney W. Grimes]

> On Sun, Mar 17, 2019 at 6:06 PM Mikael Abrahamsson  wrote:
> >
> > On Sat, 16 Mar 2019, Holland, Jake wrote:
> >
> > > Granted, it still remains to be seen whether SCE in practice can match
> > > the results of L4S, and L4S was here first.  But it seems to me L4S comes
> > > with some problems that have not yet been examined, and that are nicely
> > > dodged by a SCE-based approach.
> >
> > I'm actually not that interested in an academic competition about what
> > solution gives the ultimate "best" outcome in simulation or in a lab.
> >
> > I am interested in good enough solutions that are actually deployable and
> > will get deployed, and doesn't have any pathological behaviour when it
> > comes to legacy traffic.
> >
> > Right now the Internet is full of deep FIFOs and they're not going away,
> > and they're not getting FQ_CODEL or CAKE.
> >
> > CAKE/FQ_CODEL is nice, but it's not being deployed at the typical
> > congestion points we have in real life. These devices would have a much
> > easier time getting PIE or even RED, if it was just implemented.
> >
> 
> is there an open source implementation of PIE which is close to what
> is used by the DOCSIS modems ?

I do not know if it is close to the DOCSIS modems,
but FreeBSD has PIE implemented

/usr/src/sys/netpfil/ipfw/dn_aqm_pie.c
/usr/src/sys/netpfil/ipfw/dn_aqm_pie.h
/usr/src/sys/netpfil/ipfw/dn_sched_fq_pie.c

> > Mikael Abrahamssonemail: swm...@swm.pp.se

-- 
Rod Grimes rgri...@freebsd.org
___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Greg White]

That is an interesting point.  The goal of that MUST statement is to ensure 
per-flow-fairness.  Since fq provides per-flow-fairness as a guarantee via DRR, 
there is no need to actively apply that rule, it should just be an emergent 
property of the scheduling.

-Greg


From: "Holland, Jake" 
Date: Wednesday, March 20, 2019 at 4:38 PM
To: Greg White 
Cc: tsvwg IETF list , bloat 
Subject: Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and 
experimentation of TCP Prague/L4S hackaton at IETF104

Thanks Greg,

But wouldn’t this potentially violate at least one MUST from section 5.2 of 
l4s-id?

   The likelihood that an AQM drops a Not-ECT Classic packet (p_C) MUST
   be roughly proportional to the square of the likelihood that it would
   have marked it if it had been an L4S packet (p_L)
https://tools.ietf.org/html/draft-ietf-tsvwg-ecn-l4s-id-06#section-5.2

Maybe it depends on how far you stretch “roughly”, I guess...  I’m not sure, 
but I’d imagine some realistic conditions could provide counterexamples, unless 
there’s some reason I’m missing that prevents it?

Regards,
Jake

From: Greg White 
Date: 2019-03-20 at 14:49
To: "Holland, Jake" , Bob Briscoe , 
"David P. Reed" , Vint Cerf 
Cc: tsvwg IETF list , bloat 
Subject: Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and 
experimentation of TCP Prague/L4S hackaton at IETF104

I responded to point 2 separately.  In response to point 1, the dual queue 
mechanism is not the only way to support L4S and TCP Prague.  As we’ve 
mentioned a time or two, an fq_codel system can also support L4S and TCP 
Prague.  I’m not aware that anyone has implemented it to test it out yet 
(because so far most interest has been on dual-queue), but I think the simplest 
version is:

At dequeue:
If packet indicates ECT(1):
If sojourn_time > L4S_threshold:
Set CE*, and forward packet
Else:
Forward packet
Else:
Do all the normal CoDel stuff

In many cases, all of the packets in a single queue will either all be ECT(1) 
or none of them will.  But, to handle VPNs (where a mix of ECT(1) and 
non-ECT(1) packets could share a queue), I would think that including the ECN 
field in the flow hash would keep those packets separate.

*a more sophisticated approach would be to mark CE based on a ramp function 
between a min_thresh and max_thresh, which could be implemented as a random 
draw, or via a counting function




From: Bloat  on behalf of "Holland, Jake" 

Date: Wednesday, March 20, 2019 at 1:06 PM
To: Bob Briscoe , "David P. Reed" , 
Vint Cerf 
Cc: tsvwg IETF list , bloat 
Subject: Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and 
experimentation of TCP Prague/L4S hackaton at IETF104

Hi Bob & Greg,

I agree there has been a reasonably open conversation about the L4S
work, and thanks for all your efforts to make it so.

However, I think there's 2 senses in which "private" might be fair that
I didn't see covered in your rebuttals (merging forks and including
Greg's rebuttal by reference from here:
https://lists.bufferbloat.net/pipermail/bloat/2019-March/009038.html )

Please note:
I don't consider these senses of "private" a disqualifying argument
against the use of L4S, though I do consider them costs that should be
taken into account (and of course opinions may differ here).

With that said, I wondered whether either of you have any responses that
speak to these points:


1. the L4S use of the ECT(1) codepoint can't be marked CE except by a
patent-protected AQM scheduler.

I understand that l4s-id suggests the possibility of an alternate
scheme.  However, comparing the MUSTs of the section 5 requirements
with the claims made by the patent seems to leave no room for an
alternate that would not infringe the patent claims, unless I'm missing
something?

https://tools.ietf.org/html/draft-ietf-tsvwg-ecn-l4s-id-06#section-5
https://patents.google.com/patent/US20170019343A1/en


2. the L4S use of the ECT(1) codepoint privileges the low latency use
case.

As Jonathan Morton pointed out, low latency is only one of several
known use cases that would be worthwhile to identify to an AQM
scheduler, which the L4S approach cannot be extended to handle:
- Minimize Latency
- Minimize Loss
- Minimize Cost
- Maximize Throughput

https://lists.bufferbloat.net/pipermail/ecn-sane/2019-March/66.html

I understand that "Minimize Cost" is perhaps covered by LEPHB, and that
operator tuning parameters for a dualq node can possibly allow for
tuning between minimizing loss and latency, as mentioned in section
4.1 of aqm-dualq, but since the signaling is bundled together, it can
only happen for one at a time, if I'm reading it right.

But more importantly, the L4S usage couples the minimized latency use
case to any possibility

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Holland, Jake]

Thanks Greg,

But wouldn’t this potentially violate at least one MUST from section 5.2 of 
l4s-id?

   The likelihood that an AQM drops a Not-ECT Classic packet (p_C) MUST
   be roughly proportional to the square of the likelihood that it would
   have marked it if it had been an L4S packet (p_L)
https://tools.ietf.org/html/draft-ietf-tsvwg-ecn-l4s-id-06#section-5.2

Maybe it depends on how far you stretch “roughly”, I guess...  I’m not sure, 
but I’d imagine some realistic conditions could provide counterexamples, unless 
there’s some reason I’m missing that prevents it?

Regards,
Jake

From: Greg White 
Date: 2019-03-20 at 14:49
To: "Holland, Jake" , Bob Briscoe , 
"David P. Reed" , Vint Cerf 
Cc: tsvwg IETF list , bloat 
Subject: Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and 
experimentation of TCP Prague/L4S hackaton at IETF104

I responded to point 2 separately.  In response to point 1, the dual queue 
mechanism is not the only way to support L4S and TCP Prague.  As we’ve 
mentioned a time or two, an fq_codel system can also support L4S and TCP 
Prague.  I’m not aware that anyone has implemented it to test it out yet 
(because so far most interest has been on dual-queue), but I think the simplest 
version is:

At dequeue:
If packet indicates ECT(1):
If sojourn_time > L4S_threshold:
Set CE*, and forward packet
Else:
Forward packet
Else:
Do all the normal CoDel stuff

In many cases, all of the packets in a single queue will either all be ECT(1) 
or none of them will.  But, to handle VPNs (where a mix of ECT(1) and 
non-ECT(1) packets could share a queue), I would think that including the ECN 
field in the flow hash would keep those packets separate.

*a more sophisticated approach would be to mark CE based on a ramp function 
between a min_thresh and max_thresh, which could be implemented as a random 
draw, or via a counting function




From: Bloat  on behalf of "Holland, Jake" 

Date: Wednesday, March 20, 2019 at 1:06 PM
To: Bob Briscoe , "David P. Reed" , 
Vint Cerf 
Cc: tsvwg IETF list , bloat 
Subject: Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and 
experimentation of TCP Prague/L4S hackaton at IETF104

Hi Bob & Greg,

I agree there has been a reasonably open conversation about the L4S
work, and thanks for all your efforts to make it so.

However, I think there's 2 senses in which "private" might be fair that
I didn't see covered in your rebuttals (merging forks and including
Greg's rebuttal by reference from here:
https://lists.bufferbloat.net/pipermail/bloat/2019-March/009038.html )

Please note:
I don't consider these senses of "private" a disqualifying argument
against the use of L4S, though I do consider them costs that should be
taken into account (and of course opinions may differ here).

With that said, I wondered whether either of you have any responses that
speak to these points:


1. the L4S use of the ECT(1) codepoint can't be marked CE except by a
patent-protected AQM scheduler.

I understand that l4s-id suggests the possibility of an alternate
scheme.  However, comparing the MUSTs of the section 5 requirements
with the claims made by the patent seems to leave no room for an
alternate that would not infringe the patent claims, unless I'm missing
something?

https://tools.ietf.org/html/draft-ietf-tsvwg-ecn-l4s-id-06#section-5
https://patents.google.com/patent/US20170019343A1/en


2. the L4S use of the ECT(1) codepoint privileges the low latency use
case.

As Jonathan Morton pointed out, low latency is only one of several
known use cases that would be worthwhile to identify to an AQM
scheduler, which the L4S approach cannot be extended to handle:
- Minimize Latency
- Minimize Loss
- Minimize Cost
- Maximize Throughput

https://lists.bufferbloat.net/pipermail/ecn-sane/2019-March/66.html

I understand that "Minimize Cost" is perhaps covered by LEPHB, and that
operator tuning parameters for a dualq node can possibly allow for
tuning between minimizing loss and latency, as mentioned in section
4.1 of aqm-dualq, but since the signaling is bundled together, it can
only happen for one at a time, if I'm reading it right.

But more importantly, the L4S usage couples the minimized latency use
case to any possibility of getting a high fidelity explicit congestion
signal, so the "maximize throughput" use case can't ever get it.


Regards,
Jake

PS:
If you get a chance, I'm still interested in seeing answers to my
questions about deployment mitigations on the tsvwg list:
https://mailarchive.ietf.org/arch/msg/tsvwg/TWOVpI-SvVsYVy0_U6K8R04eq3A

I'm not surprised if it slipped by unnoticed, there have been a lot of
emails on this.  But good answers to those questions would go a long way
toward easing my concerns 

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Loganaden Velvindron]

On Sun, Mar 17, 2019 at 6:06 PM Mikael Abrahamsson  wrote:
>
> On Sat, 16 Mar 2019, Holland, Jake wrote:
>
> > Granted, it still remains to be seen whether SCE in practice can match
> > the results of L4S, and L4S was here first.  But it seems to me L4S comes
> > with some problems that have not yet been examined, and that are nicely
> > dodged by a SCE-based approach.
>
> I'm actually not that interested in an academic competition about what
> solution gives the ultimate "best" outcome in simulation or in a lab.
>
> I am interested in good enough solutions that are actually deployable and
> will get deployed, and doesn't have any pathological behaviour when it
> comes to legacy traffic.
>
> Right now the Internet is full of deep FIFOs and they're not going away,
> and they're not getting FQ_CODEL or CAKE.
>
> CAKE/FQ_CODEL is nice, but it's not being deployed at the typical
> congestion points we have in real life. These devices would have a much
> easier time getting PIE or even RED, if it was just implemented.
>

is there an open source implementation of PIE which is close to what
is used by the DOCSIS modems ?

> --
> Mikael Abrahamssonemail: swm...@swm.pp.se
> ___
> Ecn-sane mailing list
> ecn-s...@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/ecn-sane
___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Vint Cerf]

thanks, David - that's the information I was looking for.

v


On Sun, Mar 17, 2019 at 2:07 PM David P. Reed  wrote:

> Vint -
>
>
>
> BBR is the end-to-end control logic that adjusts the source rate to match
> the share of the bolttleneck link it should use.
>
>
>
> It depends on getting reliable current congestion information via packet
> drops and/or ECN.
>
>
>
> So the proposal by these guys (not the cable guys) is an attempt to
> improve the quality of the congestion signal inserted by the router with
> the bottleneck outbound link.
>
>
>
> THe cable guys are trying to get a "private" field in the IP header for
> their own use.
>
>
>
>
>
> -Original Message-
> From: "Vint Cerf" 
> Sent: Saturday, March 16, 2019 5:57pm
> To: "Holland, Jake" 
> Cc: "Mikael Abrahamsson" , "David P. Reed" <
> dpr...@deepplum.com>, "ecn-s...@lists.bufferbloat.net" <
> ecn-s...@lists.bufferbloat.net>, "bloat" 
> Subject: Re: [Ecn-sane] [Bloat] [iccrg] Fwd: [tcpPrague] Implementation
> and experimentation of TCP Prague/L4S hackaton at IETF104
>
> where does BBR fit into all this?
> v
>
> On Sat, Mar 16, 2019 at 5:39 PM Holland, Jake  wrote:
>
>> On 2019-03-15, 11:37, "Mikael Abrahamsson"  wrote:
>> L4S has a much better possibility of actually getting deployment into
>> the
>> wider Internet packet-moving equipment than anything being talked
>> about
>> here. Same with PIE as opposed to FQ_CODEL. I know it's might not be
>> as
>> good, but it fits better into actual silicon and it's being proposed
>> by
>> people who actually have better channels into the people setting hard
>> requirements.
>>
>> I suggest you consider joining them instead of opposing them.
>>
>>
>> Hi Mikael,
>>
>> I agree it makes sense that fq_anything has issues when you're talking
>> about the OLT/CMTS/BNG/etc., and I believe it when you tell me PIE
>> makes better sense there.
>>
>> But fq_x makes great sense and provides real value for the uplink in a
>> home, small office, coffee shop, etc. (if you run the final rate limit
>> on the home side of the access link.)  I'm thinking maybe there's a
>> disconnect here driven by the different use cases for where AQMs can go.
>>
>> The thing is, each of these is the most likely congestion point at
>> different times, and it's worthwhile for each of them to be able to
>> AQM (and mark packets) under congestion.
>>
>> One of the several things that bothers me with L4S is that I've seen
>> precious little concern over interfering with the ability for another
>> different AQM in-path to mark packets, and because it changes the
>> semantics of CE, you can't have both working at the same time unless
>> they both do L4S.
>>
>> SCE needs a lot of details filled in, but it's so much cleaner that it
>> seems to me there's reasonably obvious answers to all (or almost all) of
>> those detail questions, and because the semantics are so much cleaner,
>> it's much easier to tell it's non-harmful.
>>
>> 
>> The point you raised in another thread about reordering is mostly
>> well-taken, and a good counterpoint to the claim "non-harmful relative
>> to L4S".
>>
>> To me it seems sad and dumb that switches ended up trying to make
>> ordering guarantees at cost of switching performance, because if it's
>> useful to put ordering in the switch, then it must be equally useful to
>> put it in the receiver's NIC or OS.
>>
>> So why isn't it in all the receivers' NIC or OS (where it would render
>> the switch's ordering efforts moot) instead of in all the switches?
>>
>> I'm guessing the answer is a competition trap for the switch vendors,
>> plus "with ordering goes faster than without, when you benchmark the
>> switch with typical load and current (non-RACK) receivers".
>>
>> If that's the case, it seems like the drive for a competitive advantage
>> caused deployment of a packet ordering workaround in the wrong network
>> location(s), out of a pure misalignment of incentives.
>>
>> RACK rates to fix that in the end, but a lot of damage is already done,
>> and the L4S approach gives switches a flag that can double as proof that
>> RACK is there on the receiver, so they can stop trying to order those
>> packets.
>>
>> So point granted, I understand and agree there's a cost to abandoning
>> that advantage.
>> 
>>
>> But as you also said so well in another thread, this is important.  ("The
>> last unicorn", IIRC.)  How much does it matter if there's a feature that
>> has value today, but only until RACK is widely deployed?  If you were
>> convinced RACK would roll out everywhere within 3 years and SCE would
>> produce better results than L4S over the following 15 years, would that
>> change your mind?
>>
>> It would for me, and that's why I'd like to see SCE explored before
>> making a call.  I think at its core, it provides the same thing L4S does
>> (a high-fidelity explicit congestion signal for the sender), but with
>> much cleaner semantics that can be incrementally added to congestion
>> contro

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Vint Cerf]

where does BBR fit into all this?

v


On Sat, Mar 16, 2019 at 5:39 PM Holland, Jake  wrote:

> On 2019-03-15, 11:37, "Mikael Abrahamsson"  wrote:
> L4S has a much better possibility of actually getting deployment into
> the
> wider Internet packet-moving equipment than anything being talked
> about
> here. Same with PIE as opposed to FQ_CODEL. I know it's might not be
> as
> good, but it fits better into actual silicon and it's being proposed
> by
> people who actually have better channels into the people setting hard
> requirements.
>
> I suggest you consider joining them instead of opposing them.
>
>
> Hi Mikael,
>
> I agree it makes sense that fq_anything has issues when you're talking
> about the OLT/CMTS/BNG/etc., and I believe it when you tell me PIE
> makes better sense there.
>
> But fq_x makes great sense and provides real value for the uplink in a
> home, small office, coffee shop, etc. (if you run the final rate limit
> on the home side of the access link.)  I'm thinking maybe there's a
> disconnect here driven by the different use cases for where AQMs can go.
>
> The thing is, each of these is the most likely congestion point at
> different times, and it's worthwhile for each of them to be able to
> AQM (and mark packets) under congestion.
>
> One of the several things that bothers me with L4S is that I've seen
> precious little concern over interfering with the ability for another
> different AQM in-path to mark packets, and because it changes the
> semantics of CE, you can't have both working at the same time unless
> they both do L4S.
>
> SCE needs a lot of details filled in, but it's so much cleaner that it
> seems to me there's reasonably obvious answers to all (or almost all) of
> those detail questions, and because the semantics are so much cleaner,
> it's much easier to tell it's non-harmful.
>
> 
> The point you raised in another thread about reordering is mostly
> well-taken, and a good counterpoint to the claim "non-harmful relative
> to L4S".
>
> To me it seems sad and dumb that switches ended up trying to make
> ordering guarantees at cost of switching performance, because if it's
> useful to put ordering in the switch, then it must be equally useful to
> put it in the receiver's NIC or OS.
>
> So why isn't it in all the receivers' NIC or OS (where it would render
> the switch's ordering efforts moot) instead of in all the switches?
>
> I'm guessing the answer is a competition trap for the switch vendors,
> plus "with ordering goes faster than without, when you benchmark the
> switch with typical load and current (non-RACK) receivers".
>
> If that's the case, it seems like the drive for a competitive advantage
> caused deployment of a packet ordering workaround in the wrong network
> location(s), out of a pure misalignment of incentives.
>
> RACK rates to fix that in the end, but a lot of damage is already done,
> and the L4S approach gives switches a flag that can double as proof that
> RACK is there on the receiver, so they can stop trying to order those
> packets.
>
> So point granted, I understand and agree there's a cost to abandoning
> that advantage.
> 
>
> But as you also said so well in another thread, this is important.  ("The
> last unicorn", IIRC.)  How much does it matter if there's a feature that
> has value today, but only until RACK is widely deployed?  If you were
> convinced RACK would roll out everywhere within 3 years and SCE would
> produce better results than L4S over the following 15 years, would that
> change your mind?
>
> It would for me, and that's why I'd like to see SCE explored before
> making a call.  I think at its core, it provides the same thing L4S does
> (a high-fidelity explicit congestion signal for the sender), but with
> much cleaner semantics that can be incrementally added to congestion
> controls that people are already using.
>
> Granted, it still remains to be seen whether SCE in practice can match
> the results of L4S, and L4S was here first.  But it seems to me L4S comes
> with some problems that have not yet been examined, and that are nicely
> dodged by a SCE-based approach.
>
> If L4S really is as good as they seem to think, I could imagine getting
> behind it, but I don't think that's proven yet.  I'm not certain, but
> all the comparative analyses I remember seeing have been from more or
> less the same team, and I'm not convinced they don't have some
> misaligned incentives of their own.
>
> I understand a lot of work has gone into L4S, but this move to jump it
> from interesting experiment to de-facto standard without a more critical
> review that digs deeper into some of the potential deployment problems
> has me concerned.
>
> If it really does turn out to be good enough to be permanent, I'm not
> opposed to it, but I'm just not convinced that it's non-harmful, and my
> default position is that the cleaner solution is going to be better in
> the long run, if they can do the same job.
>
> It's not t

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Bob Briscoe]

Roland,

On 22/03/2019 13:53, Bless, Roland (TM) wrote:

Hi Bob,

see inline.

Am 21.03.19 um 14:24 schrieb Bob Briscoe:

On 21/03/2019 08:49, Bless, Roland (TM) wrote:

Hi,

Am 21.03.19 um 09:02 schrieb Bob Briscoe:

Just to rapidly reply,


On 21/03/2019 07:46, Jonathan Morton wrote:

The ECN field was never intended to be used as a classifier, except to
distinguish Not-ECT flows from ECT flows (which a middlebox does need
to know, to choose between mark and drop behaviours).  It was intended
to be used to convey congestion information from the network to the
receiver.  SCE adheres to that ideal.

Each PHB has a forwarding behaviour a DSCP re-marking behaviour and an
ECN marking behaviour. The ECN field is the claissifer for the ECN
marking behaviour.

That's exactly the reason, why using ECT(1) as classifier for L4S
behavior is not the right choice. L4S should use a DSCP for
classification, because it is actually defining a PHB.

1/ First Terminology
The definition of 'PHB' includes the drop or ECN-marking behaviour. For
instance, you see this in WRED or in PCN (Pre-Congestion Notification).
If you want to solely talk about scheduling, pls say the scheduling PHB.

I thought that I'm well versed with Diffserv terminology, but I'm not
aware that a Diffserv PHB requires the definition of an ECN marking
behavior.
Ah well, I do not think that any living human has visited all the dark 
corners of the Diffserv world.


I didn't mean (or say) that a Diffserv PHB /requires/ an ECN marking 
behaviour, just that if you are talking about a Diffserv PHB that 
includes an ECN marking behaviour, it helps to say when you are solely 
talking about the scheduling part of the PHB.


In many cases when there is no ECN marking behaviour, it makes no 
difference if you omit the word scheduling, cos that is all there is to 
the behaviour.



In fact ECN is orthogonal to Diffserv as both RFCs 2474 and
2475 do not even mention ECN. RFC 2475:
"A per-hop behavior (PHB) is a description of the externally
observable forwarding behavior of a DS node applied to a particular
DS behavior aggregate." and "Useful behavioral distinctions
are mainly observed when multiple behavior aggregates compete for
buffer and bandwidth resources on a node."
Even the original experimental ECN spec RFC2481 was published just after 
2474 & 2475. So you wouldn't expect the original Diffserv specs to 
mention something that didn't exist then.




Usually, there are different mechanisms how to implement a PHB,
e.g., for EF one could use a tail drop queue and Simple Priority
Queueing, Weighted Fair Queueing, or Deficit Round Robin and so
on. Consequently, queueing and scheduling behavior are used to
_implement_ a PHB, i.e., IMHO it makes sense to distinguish between
the PHB as externally observable behavior and a specific _PHB
implementation_ as also pointed out in RFC2475:
PHBs are implemented in nodes by means of some buffer management and
packet scheduling mechanisms.  PHBs are defined in terms of behavior
characteristics relevant to service provisioning policies, and not in
terms of particular implementation mechanisms.


So some of the Diffserv PHBs do _not_ require using an AQM,
which is often the basis for ECN marking, e.g., for EF
tail drop should be sufficient. For other PHBs it may be
useful to say something about ECN usage (as I did for LE).

RFC 2475:

PHBs may be specified in terms of their resource (e.g., buffer,
bandwidth) priority relative to other PHBs, or in terms of their
relative observable traffic characteristics (e.g., delay, loss).
Since RFC2474 & 2475, AQM behaviour and/or ECN marking behaviour has 
become part of some Diffserv PHBs. E.g. WRED in AF. See any of the 
tables in RFC4594 that have an AQM column.


The need for ECN marking behaviour (rather than just AQM behaviour in 
general) as part of a PHB became necessary during the definition of PCN. 
Jo Babiarz and Kwok Ho Chan were both authors of 4594 and of many of the 
PCN specs, and proposed the term 'marking behaviour' as part of the PHB. 
You will find ECN marking behaviours are central to, for instance, RFC5670.


As I've pointed out already, the transitions used by SCE were already in 
the PCN baseline encoding spec [RFC6660], except only defined if 
accompanied by a specific DSCP (which was subsequently standardized as 
EF-ADMIT).





I think that L4S therefore specifies such a PHB as it is defined
in relation to the default PHB (as in the L4S arch draft
"Classic service").
No. The L4S use of ECN is orthogonal to Diffserv, and can be associated 
with more than one scheduling PHB in a queuing hierarchy. See 
draft-briscoe-l4s-diffserv (which I would welcome you to review - Brian 
Carpenter is also currently reviewing it).


However, you are certainly right in thinking that L4S associated with 
the default PHB is by far and away the most important use-case for L4S. 
All the other possible schemes in l4s-diffserv are only possibilities - 
probabl

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Bob Briscoe]

Alex, inline...

On 24/03/2019 21:15, alex.b...@ealdwulf.org.uk wrote:


Hi Bob,


I note that all the non-dependent claims of US20170019343A1 (claims 1,14,22) 
seem to assume use of the proportional-integral controller (Note, I am not a 
lawyer, and especially not a patent lawyer).
Yes, as I understand it, Nokia's intention with this filing was to cover 
use of the PI controller in particular, in combination with various 
other ideas.



In Appendix B of draft-briscoe-tsvwg-aqm-dualq-coupled, an alternate algorithm 
'Curvy RED' seems to replace PI, but it is noted that 'the Curvy RED algorithm 
has not been maintained to the same degree as the DualPI2 algorithm '.

Can you comment on whether the Curvy RED algorithm could form a 
non-patent-encumbered dualq? In particular:
  - Why wasn't curvy red further developed? Was it found to contain some 
deficiency? Are you intending to present it as an alternative?
We just didn't develop it further, cos we were getting better results 
with PI2. However, I am aware of a hardware implementation based on 
Curvy RED going on at the moment, and you will see there have recently 
been review comments on that Curvy RED appendix on the list.


So, even tho PI might be better, Curvy RED (or another AQM) might be 
preferable for other reasons that performance (e.g. ease of 
implementation, or similarity to an existing hardware implementation).


And indeed, there's nothing to stop anyone using other AQMs, either to 
work round the IPR, or because they're preferable in their own right - 
the DualQ Coupled AQM is intentionally a framework into which you drop 2 
AQMs.



  - Does Curvy RED actually completely replace PI?

Yes.

  - Can we have reasonable assurance that no patents will surface covering 
Curvy RED?
Well, I developed the idea of Curvy RED and I / my employer (BT) did not 
file any IPR at the time. I got approval to publish a tech report 
jointly with Al-Lu. http://bobbriscoe.net/pubs.html#CRED-insights


That was May 2015, so given nothing has surfaced by now, there can't be 
anything from that time from us (where us = Al-Lu & BT).


Of course, I cannot guarantee that there is not another patent in the 
system from some other random company that my searches haven't found. 
There are large numbers of AQM patents. Also, I cannot guarantee that an 
implementer working now isn't filing patents around their 
implementation. All we can do is publish as much as possible as early as 
possible to try to keep some areas of the field open.



Bob


Thanks,
Alex


On Wednesday, March 20, 2019, 11:29:38 PM GMT, Bob Briscoe 
 wrote:






1/ In 2016, I arranged for the hire of a patent attorney to undertake the 
unusual step of filing a third party observation with the European Patent 
Office. This went through Al-Lu's patent application claim by claim pointing to 
prior art and giving the patent examiner all the arguments to reject each 
claim. However, the examiner chose to take very little note of it, which was 
disappointing and costly for us. The main prior art is:
     Gibbens, R.J. & Kelly, F.P., "On Packet Marking at Priority Queues," IEEE 
Transactions on Automatic Control 47(6):1016--1020 (June 2002)
The guys named as inventors in AL-Lu's filing published a paper on PI2 with me, 
in which we included a citation of this Gibbens paper as inspiration for the 
coupling. The Gibbens paper was already cited as background by other patents, 
so the EPO has it in their citation index.

The coupling was also based on my prior research with Mirja before I started 
working with the guys from Al-Lu in the RITE European Collaborative project. we 
had to go through a few rejections, but Mirja and I finally got this work 
published in 2014  - still before the priority date of the Al-Lu patent 
application:
     Kühlewind, M., Wagner, D.P., Espinosa, J.M.R. & Briscoe, B., "Using Data Center 
TCP (DCTCP) in the Internet," In: Proc. Third IEEE Globecom Workshop on 
Telecommunications Standards: From Research to Standards pp.583-588 (December 2014)

2/ The only claim that I could not find prior art for (in the original EU 
filing) was a very specific claim about using a square root for the coupling. 
The Linux implementation runs this the other way round so that it only has to 
do a squaring. So I figured we were safe from that.

However, until just now, I had not noticed that Al-Lu has retrospectively 
re-written the claims in the US patent and in the EU patent application to 
claim this the other way round - as a squaring. And to claim the two random 
number trick. Both restructuring to use a squaring and the two random number 
trick were definitely my ideas (while working for BT in a collaboration with 
Al-Lu). I have emails to prove this (from memory they were actually both in the 
same email). This is important, because a patent has to be about mechanism, not 
algorithm.

3/ This is a positive development. It means this patent is on very shaky legal 
ground. I have been trying to p

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Luca Muscariello]

For enterprise networks interconnection to the cloud (public or private)
can be done with e2e DSCP marking.
This is just part of products available at AWS/Azure w/ or w/o their own
interconnection network (direct connect, express route)
and used regularly to ameliorate QoS of RTC applications such as WebEx,
Skype for business, WebRTC.
As I said, peering is key and cloud providers are doing that very well.
With the current widespread of SaaS and IaaS your app is likely running in
the cloud.

Even in case a branch office is connected using SD-WAN, DSCP is tunnelled
by it, and transparently for the application.

If working from home, the quality of peering of an SP to cloud providers is
key.
I'd say that the Cloud and the decline of transit networks have facilitated
DSCP e2e.




On Sat, Mar 23, 2019 at 7:40 PM Mikael Abrahamsson  wrote:

> On Sat, 23 Mar 2019, Luca Muscariello wrote:
>
> > It the app runs in the cloud and the cloud has direct peering links to
> > your branch office or SP most likely DSCP works.
>
> Do you have numbers to back this statement up? In my experience direct
> peering links has nothing to do with this, instead remaking is done
> equally at the customer edge and peering/transit edge respectively.
>
> --
> Mikael Abrahamssonemail: swm...@swm.pp.se
>
___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Roland Bless]

Hi Michael,

thanks for the pointer. Looking forward to the talk...

Regards
 Roland


On 23.03.19 at 18:48 Michael Welzl wrote:
> Hi,
> 
> I’ll do what academics do and add our own data point, below:
> 
>> On Mar 23, 2019, at 4:19 PM, Roland Bless > > wrote:
>>
>> Hi Mikael,
>>
>> On 23.03.19 at 11:02 Mikael Abrahamsson wrote:
>>> On Sat, 23 Mar 2019, Roland Bless wrote:
>>>
 I suggest to use an additional DSCP to mark L4S packets.
>>>
>>> DSCP doesn't work end-to-end on the Internet, so what you're suggesting
>>> isn't a workable solution.
>>
>> It's true that DSCPs may be remarked, but RFC 2474
>> already stated
>>
>>   Packets received with an unrecognized codepoint SHOULD be forwarded
>>   as if they were marked for the Default behavior (see Sec. 4), and
>>   their codepoints should not be changed.
>>
>> The rtcweb WG also counts on e2e DSCPs.
>> After the LE PHB experience I would suggest to probably use
>> DSCP 5 which has got better chances to survive ToS bleaching (maybe
>> around 80%), if I understand
>> https://www.sciencedirect.com/science/article/pii/S0140366417312835
>> correctly. Diffserv behavior is usually configurable and can be changed
>> without replacing the network gear.
> 
> Runa Barik, Michael Welzl, Ahmed Mustafa Elmokashfi, Thomas Dreibholz,
> Stein Gjessing: "Can WebRTC QoS Work? A DSCP Measurement Study", 30th
> International Teletraffic Congress (ITC 30), 3 - 7 September 2018,
> Vienna, Austria.
> https://itc-conference.org/_Resources/Persistent/780df4482d0fe80f6180f523ebb9482c6869e98b/Barik18ITC30.pdf
> 
> We didn’t measure undefined codepoints though, only EF, AF42 and CS1.
> Table 2 compares bleaching for these codepoints with the results in the
> paper you point to; it’s quite similar.
> Well yes, there’s a significant amount of bleaching, we can’t deny that.
> But sometimes the codepoint survives, and it seems to survive
> surprisingly long into the path (fig.4 in our paper).
> 
> In the MAPRG session in Prague, Runa Barik will give a talk about the
> measured delay impact of such opportunistic use of these DSCP values by
> an end system.
> 
> Cheers,
> Michael
> 

___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Roland Bless]

Hi Jonathan,

On 23.03.19 at 16:03 Jonathan Morton wrote:
>> On 23 Mar, 2019, at 11:02 am, Mikael Abrahamsson  wrote:
>>
>> On Sat, 23 Mar 2019, Roland Bless wrote:
>>
>>> I suggest to use an additional DSCP to mark L4S packets.
>>
>> DSCP doesn't work end-to-end on the Internet, so what you're suggesting 
>> isn't a workable solution.
> 
> An interesting question, in this context, is "precisely what happens if the 
> DSCP is lost?"
> 
> With a notional L4S using DSCP instead of ECT(1) as an identifier, that 
> really is a serious problem; the L4S flow will flood out TCP-friendly flows 
> *unless* its failsafe kicks in.

One possibility to avoid such problems could be to check for DSCP
remarking on connection setup (i.e., during TCP handshake) and then fall
back to classic behavior in case that the DSCP was modified.

> But with SCE, what happens is that the L4S flow gets treated the same as any 
> other, if the bottleneck where the distinction matters is downstream of the 
> DSCP corruption.  The L4S flow reacts properly to CE in this scenario, 
> because its been designed around SCE semantics, so it is TCP-friendly.  
> Flow-isolating AQMs don't need to know the DSCP in the first place.
> 
> So the worst case is when a single or dual-queue AQM bottleneck is involved, 
> and the L4S flow is affected by queuing induced by other flows who aren't 
> quite so enlightened.  The damage is only to the L4S flow, and within 
> reasonable bounds.
> 
> This of course ignores the overwhelmingly most common situation on today's 
> Internet, where the bottleneck queue is completely unmanaged.  But then, 
> losing the DSCP has no effect there.

Regards
 Roland


___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Roland Bless]

Hi Mikael,

On 23.03.19 at 18:16 Mikael Abrahamsson wrote:
> On Sat, 23 Mar 2019, Roland Bless wrote:
> 
>> It's true that DSCPs may be remarked, but RFC 2474
>> already stated
>>
>>   Packets received with an unrecognized codepoint SHOULD be forwarded
>>   as if they were marked for the Default behavior (see Sec. 4), and
>>   their codepoints should not be changed.
> 
> https://mailman.nanog.org/pipermail/nanog/2015-May/075004.html
> 
> https://www.nanog.org/mailinglist/mailarchives/old_archive/2005-05/msg00654.html

This is pretty sad. The correct answer to the first question
"does Internet trust IP DSCP marking?" should have been twofold:
a) don't trust already present markings on ingress
  for your own supported PHBs (except default and LE PHBs :-)
  unless you have agreed with the neighboring
  DS domain.
b) Packets received with an unrecognized DSCP SHOULD be forwarded
   as best effort and their DSCP should NOT be changed.

The BCP to unconditionally bleach (set to 0) is IMHO simply wrong: one
has to distinguish between treating as default PHB and overwriting the
DSCP. For internally supported DSCPs/PHBs one typically needs to bleach
(but e.g., not for LE), but for all unsupported DSCPs simply map them to
the default PHB.
It's true that Diffserv's major line of defense is the domain
boundary that needs to protect the domain's resources against
unauthorized use. So a domain that internally supports EF should not
honor incoming EF marked packets from untrusted/unadmitted sources, and
therefore must bleach them. For unsupported DSCPs though,
one could simply _map_ them to the default PHB while retaining the DSCP.

> Please note the dates, as in 4 and 14 years ago respectively.
> 
> So please read those threads and then tell me that what you quoted above
> has bearing on reality.

It's clear that just setting everything to DSCP 0 is the safe option
(in case one has no full control over all equipment etc.),
but it has the mentioned drawback of limiting the future extensibility.
Since Diffserv requires a configurable mapping of DSCP to PHB
a consistent configuration should be possible, nevertheless.

Regards
 Roland
___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Michael Welzl]

Our paper has some data - again, fig. 4. That’s an AS level analysis, for the 
data that we had (which wasn’t huge - a couple thousand paths). For the 
majority of measurements, the DSCP survived more than 1 AS hop; in case of 
IPv6, the majority survived more than 2.

Cheers,
Michael


> On Mar 23, 2019, at 7:40 PM, Mikael Abrahamsson  wrote:
> 
> On Sat, 23 Mar 2019, Luca Muscariello wrote:
> 
>> It the app runs in the cloud and the cloud has direct peering links to your 
>> branch office or SP most likely DSCP works.
> 
> Do you have numbers to back this statement up? In my experience direct 
> peering links has nothing to do with this, instead remaking is done equally 
> at the customer edge and peering/transit edge respectively.
> 
> -- 
> Mikael Abrahamssonemail: swm...@swm.pp.se

___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Mikael Abrahamsson]

On Sat, 23 Mar 2019, Luca Muscariello wrote:

It the app runs in the cloud and the cloud has direct peering links to 
your branch office or SP most likely DSCP works.


Do you have numbers to back this statement up? In my experience direct 
peering links has nothing to do with this, instead remaking is done 
equally at the customer edge and peering/transit edge respectively.


--
Mikael Abrahamssonemail: swm...@swm.pp.se
___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Luca Muscariello]

WebEx, WebRTC they use it.
QoS works. To answer the question in the title of Michael’s paper.

It the app runs in the cloud and the cloud has direct peering links to your
branch office or SP
most likely DSCP works.

And going back to Roland’s proposal, it seems the most natural approach
instead of hacking the semantics.


On Sat 23 Mar 2019 at 18:48, Michael Welzl  wrote:

> Hi,
>
> I’ll do what academics do and add our own data point, below:
>
> On Mar 23, 2019, at 4:19 PM, Roland Bless  wrote:
>
> Hi Mikael,
>
> On 23.03.19 at 11:02 Mikael Abrahamsson wrote:
>
> On Sat, 23 Mar 2019, Roland Bless wrote:
>
> I suggest to use an additional DSCP to mark L4S packets.
>
>
> DSCP doesn't work end-to-end on the Internet, so what you're suggesting
> isn't a workable solution.
>
>
> It's true that DSCPs may be remarked, but RFC 2474
> already stated
>
>   Packets received with an unrecognized codepoint SHOULD be forwarded
>   as if they were marked for the Default behavior (see Sec. 4), and
>   their codepoints should not be changed.
>
> The rtcweb WG also counts on e2e DSCPs.
> After the LE PHB experience I would suggest to probably use
> DSCP 5 which has got better chances to survive ToS bleaching (maybe
> around 80%), if I understand
> https://www.sciencedirect.com/science/article/pii/S0140366417312835
> correctly. Diffserv behavior is usually configurable and can be changed
> without replacing the network gear.
>
>
> Runa Barik, Michael Welzl, Ahmed Mustafa Elmokashfi, Thomas Dreibholz,
> Stein Gjessing: "Can WebRTC QoS Work? A DSCP Measurement Study", 30th
> International Teletraffic Congress (ITC 30), 3 - 7 September 2018, Vienna,
> Austria.
>
> https://itc-conference.org/_Resources/Persistent/780df4482d0fe80f6180f523ebb9482c6869e98b/Barik18ITC30.pdf
>
> We didn’t measure undefined codepoints though, only EF, AF42 and CS1.
> Table 2 compares bleaching for these codepoints with the results in the
> paper you point to; it’s quite similar.
> Well yes, there’s a significant amount of bleaching, we can’t deny that.
> But sometimes the codepoint survives, and it seems to survive surprisingly
> long into the path (fig.4 in our paper).
>
> In the MAPRG session in Prague, Runa Barik will give a talk about the
> measured delay impact of such opportunistic use of these DSCP values by an
> end system.
>
> Cheers,
> Michael
>
> ___
> Bloat mailing list
> Bloat@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/bloat
>
___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Michael Welzl]

Hi,

I’ll do what academics do and add our own data point, below:

> On Mar 23, 2019, at 4:19 PM, Roland Bless  wrote:
> 
> Hi Mikael,
> 
> On 23.03.19 at 11:02 Mikael Abrahamsson wrote:
>> On Sat, 23 Mar 2019, Roland Bless wrote:
>> 
>>> I suggest to use an additional DSCP to mark L4S packets.
>> 
>> DSCP doesn't work end-to-end on the Internet, so what you're suggesting
>> isn't a workable solution.
> 
> It's true that DSCPs may be remarked, but RFC 2474
> already stated
> 
>   Packets received with an unrecognized codepoint SHOULD be forwarded
>   as if they were marked for the Default behavior (see Sec. 4), and
>   their codepoints should not be changed.
> 
> The rtcweb WG also counts on e2e DSCPs.
> After the LE PHB experience I would suggest to probably use
> DSCP 5 which has got better chances to survive ToS bleaching (maybe
> around 80%), if I understand
> https://www.sciencedirect.com/science/article/pii/S0140366417312835
> correctly. Diffserv behavior is usually configurable and can be changed
> without replacing the network gear.

Runa Barik, Michael Welzl, Ahmed Mustafa Elmokashfi, Thomas Dreibholz, Stein 
Gjessing: "Can WebRTC QoS Work? A DSCP Measurement Study", 30th International 
Teletraffic Congress (ITC 30), 3 - 7 September 2018, Vienna, Austria.
https://itc-conference.org/_Resources/Persistent/780df4482d0fe80f6180f523ebb9482c6869e98b/Barik18ITC30.pdf
 


We didn’t measure undefined codepoints though, only EF, AF42 and CS1. Table 2 
compares bleaching for these codepoints with the results in the paper you point 
to; it’s quite similar.
Well yes, there’s a significant amount of bleaching, we can’t deny that. But 
sometimes the codepoint survives, and it seems to survive surprisingly long 
into the path (fig.4 in our paper).

In the MAPRG session in Prague, Runa Barik will give a talk about the measured 
delay impact of such opportunistic use of these DSCP values by an end system.

Cheers,
Michael

___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Mikael Abrahamsson]

On Sat, 23 Mar 2019, Roland Bless wrote:


It's true that DSCPs may be remarked, but RFC 2474
already stated

  Packets received with an unrecognized codepoint SHOULD be forwarded
  as if they were marked for the Default behavior (see Sec. 4), and
  their codepoints should not be changed.


https://mailman.nanog.org/pipermail/nanog/2015-May/075004.html

https://www.nanog.org/mailinglist/mailarchives/old_archive/2005-05/msg00654.html

Please note the dates, as in 4 and 14 years ago respectively.

So please read those threads and then tell me that what you quoted above 
has bearing on reality.


--
Mikael Abrahamssonemail: swm...@swm.pp.se
___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Roland Bless]

Hi Mikael,

On 23.03.19 at 11:02 Mikael Abrahamsson wrote:
> On Sat, 23 Mar 2019, Roland Bless wrote:
> 
>> I suggest to use an additional DSCP to mark L4S packets.
> 
> DSCP doesn't work end-to-end on the Internet, so what you're suggesting
> isn't a workable solution.

It's true that DSCPs may be remarked, but RFC 2474
already stated

   Packets received with an unrecognized codepoint SHOULD be forwarded
   as if they were marked for the Default behavior (see Sec. 4), and
   their codepoints should not be changed.

The rtcweb WG also counts on e2e DSCPs.
After the LE PHB experience I would suggest to probably use
DSCP 5 which has got better chances to survive ToS bleaching (maybe
around 80%), if I understand
https://www.sciencedirect.com/science/article/pii/S0140366417312835
correctly. Diffserv behavior is usually configurable and can be changed
without replacing the network gear.


Regards
 Roland


___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Jonathan Morton]

> On 23 Mar, 2019, at 11:02 am, Mikael Abrahamsson  wrote:
> 
> On Sat, 23 Mar 2019, Roland Bless wrote:
> 
>> I suggest to use an additional DSCP to mark L4S packets.
> 
> DSCP doesn't work end-to-end on the Internet, so what you're suggesting isn't 
> a workable solution.

An interesting question, in this context, is "precisely what happens if the 
DSCP is lost?"

With a notional L4S using DSCP instead of ECT(1) as an identifier, that really 
is a serious problem; the L4S flow will flood out TCP-friendly flows *unless* 
its failsafe kicks in.

But with SCE, what happens is that the L4S flow gets treated the same as any 
other, if the bottleneck where the distinction matters is downstream of the 
DSCP corruption.  The L4S flow reacts properly to CE in this scenario, because 
its been designed around SCE semantics, so it is TCP-friendly.  Flow-isolating 
AQMs don't need to know the DSCP in the first place.

So the worst case is when a single or dual-queue AQM bottleneck is involved, 
and the L4S flow is affected by queuing induced by other flows who aren't quite 
so enlightened.  The damage is only to the L4S flow, and within reasonable 
bounds.

This of course ignores the overwhelmingly most common situation on today's 
Internet, where the bottleneck queue is completely unmanaged.  But then, losing 
the DSCP has no effect there.

 - Jonathan Morton

___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Mikael Abrahamsson]

On Sat, 23 Mar 2019, Roland Bless wrote:


I suggest to use an additional DSCP to mark L4S packets.


DSCP doesn't work end-to-end on the Internet, so what you're suggesting 
isn't a workable solution.


--
Mikael Abrahamssonemail: swm...@swm.pp.se
___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Luca Muscariello]

+1

On Sat, Mar 23, 2019 at 9:02 AM Roland Bless  wrote:

> Hi,
>
> On 22.03.19 at 19:28 Victor Hou wrote:
>
> > Broadcom has been deeply involved in developing the Low Latency DOCSIS
> > cable industry specification that Bob Briscoe mentioned.  We concur with
> > the L4S use of ECT(1).  L4S can be implemented either in a dual-queue or
> > an fq implementation. SCE cannot be implemented with a dual-queue
> > implementation; the only way to support it is with an fq
> > implementation.  An fq implementation is incompatible with the Low
> > Latency DOCSIS specification developed within the cable industry.
>
> I don't understand your rationale here.
> The basic SCE concept could be used for L4S as well.
> I suggest to use an additional DSCP to mark L4S packets.
> The L4S sender/receiver can simply react to the SCE
> markings the same way that they now react to CE, with
> the difference that it's safer to react to SCE, because
> the signal is unambiguous, whereas CE would be ambiguous
> (could be set by either classic AQM/ECN node or by
> an L4S node).
>
> Regards
>  Roland
> ___
> Bloat mailing list
> Bloat@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/bloat
>
___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Roland Bless]

Hi,

On 22.03.19 at 19:28 Victor Hou wrote:

> Broadcom has been deeply involved in developing the Low Latency DOCSIS
> cable industry specification that Bob Briscoe mentioned.  We concur with
> the L4S use of ECT(1).  L4S can be implemented either in a dual-queue or
> an fq implementation. SCE cannot be implemented with a dual-queue
> implementation; the only way to support it is with an fq
> implementation.  An fq implementation is incompatible with the Low
> Latency DOCSIS specification developed within the cable industry.

I don't understand your rationale here.
The basic SCE concept could be used for L4S as well.
I suggest to use an additional DSCP to mark L4S packets.
The L4S sender/receiver can simply react to the SCE
markings the same way that they now react to CE, with
the difference that it's safer to react to SCE, because
the signal is unambiguous, whereas CE would be ambiguous
(could be set by either classic AQM/ECN node or by
an L4S node).

Regards
 Roland
___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Bless, Roland (TM)]

Hi Bob,

see inline.

Am 21.03.19 um 14:24 schrieb Bob Briscoe:
> On 21/03/2019 08:49, Bless, Roland (TM) wrote:
>> Hi,
>>
>> Am 21.03.19 um 09:02 schrieb Bob Briscoe:
>>> Just to rapidly reply,
>>>
>>>
>>> On 21/03/2019 07:46, Jonathan Morton wrote:
 The ECN field was never intended to be used as a classifier, except to
 distinguish Not-ECT flows from ECT flows (which a middlebox does need
 to know, to choose between mark and drop behaviours).  It was intended
 to be used to convey congestion information from the network to the
 receiver.  SCE adheres to that ideal.
>>> Each PHB has a forwarding behaviour a DSCP re-marking behaviour and an
>>> ECN marking behaviour. The ECN field is the claissifer for the ECN
>>> marking behaviour.
>> That's exactly the reason, why using ECT(1) as classifier for L4S
>> behavior is not the right choice. L4S should use a DSCP for
>> classification, because it is actually defining a PHB.
> 
> 1/ First Terminology
> The definition of 'PHB' includes the drop or ECN-marking behaviour. For
> instance, you see this in WRED or in PCN (Pre-Congestion Notification). 
> If you want to solely talk about scheduling, pls say the scheduling PHB.

I thought that I'm well versed with Diffserv terminology, but I'm not
aware that a Diffserv PHB requires the definition of an ECN marking
behavior. In fact ECN is orthogonal to Diffserv as both RFCs 2474 and
2475 do not even mention ECN. RFC 2475:
"A per-hop behavior (PHB) is a description of the externally
observable forwarding behavior of a DS node applied to a particular
DS behavior aggregate." and "Useful behavioral distinctions
are mainly observed when multiple behavior aggregates compete for
buffer and bandwidth resources on a node."

Usually, there are different mechanisms how to implement a PHB,
e.g., for EF one could use a tail drop queue and Simple Priority
Queueing, Weighted Fair Queueing, or Deficit Round Robin and so
on. Consequently, queueing and scheduling behavior are used to
_implement_ a PHB, i.e., IMHO it makes sense to distinguish between
the PHB as externally observable behavior and a specific _PHB
implementation_ as also pointed out in RFC2475:
   PHBs are implemented in nodes by means of some buffer management and
   packet scheduling mechanisms.  PHBs are defined in terms of behavior
   characteristics relevant to service provisioning policies, and not in
   terms of particular implementation mechanisms.


So some of the Diffserv PHBs do _not_ require using an AQM,
which is often the basis for ECN marking, e.g., for EF
tail drop should be sufficient. For other PHBs it may be
useful to say something about ECN usage (as I did for LE).

RFC 2475:

   PHBs may be specified in terms of their resource (e.g., buffer,
   bandwidth) priority relative to other PHBs, or in terms of their
   relative observable traffic characteristics (e.g., delay, loss).

I think that L4S therefore specifies such a PHB as it is defined
in relation to the default PHB (as in the L4S arch draft
"Classic service").

> 2/ The architectural intent of the ECN field
> 
> For many years (long before we thought of L4S) I have been making sure
> that ECN propagation through the layers supports the duality of ECN
> behaviours as both a classifier (on the way down from L7/L4 to L3/2) and
> as a return value (on the way back up).
> 
> The architecture of ECN is determined by the valid codepoint
> transitions. They are:

I wouldn't say that it's determined solely by the transitions.

> 1. 00->11
> 2. 10->11
> 3. 01->11
> 4. 10->01
> 
> The first three were in RFC3168, but it did not preclude the fourth.
> The fourth was first standardized in RFC6660 (which I co-authored). This
> had to be isolated from the e2e use of ECN by inclusion of a DSCP as well.
> 
> The relatively late addition of the fourth approach means that an
> attempt to mark using the SCE approach (10->01) is more likely to find
> that it gets reversed when the outer header is decapsulated, if the
> decapsulator hasn't been updated to the latest RFC that catered for this
> fourth transition (RFC6040, also co-authored by me).
> 
> L4S follows the original RFC3168 approach
> SCE uses the fourth
> 
> So, SCE proposes to use /a/ correct approach, but it might not work.

In case of nodes that implement RFC6040? I think that it would
be useful to measure how many boxes out there actually do this
(or how widespread is ECN usage actually, e.g., how many boxes
actually set CE on congestion? MAMI results anyone?).

> Whereas L4S uses the original correct approach.

Which might also not work...in case RFC3168 boxes set CE, so
the L4S receivers/senders cannot know that the CE wasn't set
by an L4S node.

> 3a/ DualQ L4S AQMs
> With the DualQ, the difference between the two queues is both in their
> ECN marking behaviour and in their forwarding/scheduling behaviour.
> However, whenever there's traffic in the classic queue the coupling
> between the AQMs overrides the network scheduler. The cou

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Bob Briscoe]

Roland,

On 21/03/2019 08:49, Bless, Roland (TM) wrote:

Hi,

Am 21.03.19 um 09:02 schrieb Bob Briscoe:

Just to rapidly reply,


On 21/03/2019 07:46, Jonathan Morton wrote:

The ECN field was never intended to be used as a classifier, except to
distinguish Not-ECT flows from ECT flows (which a middlebox does need
to know, to choose between mark and drop behaviours).  It was intended
to be used to convey congestion information from the network to the
receiver.  SCE adheres to that ideal.

Each PHB has a forwarding behaviour a DSCP re-marking behaviour and an
ECN marking behaviour. The ECN field is the claissifer for the ECN
marking behaviour.

That's exactly the reason, why using ECT(1) as classifier for L4S
behavior is not the right choice. L4S should use a DSCP for
classification, because it is actually defining a PHB.


1/ First Terminology
The definition of 'PHB' includes the drop or ECN-marking behaviour. For 
instance, you see this in WRED or in PCN (Pre-Congestion Notification).  
If you want to solely talk about scheduling, pls say the scheduling PHB.


2/ The architectural intent of the ECN field

For many years (long before we thought of L4S) I have been making sure 
that ECN propagation through the layers supports the duality of ECN 
behaviours as both a classifier (on the way down from L7/L4 to L3/2) and 
as a return value (on the way back up).


The architecture of ECN is determined by the valid codepoint 
transitions. They are:

1. 00->11
2. 10->11
3. 01->11
4. 10->01

The first three were in RFC3168, but it did not preclude the fourth.
The fourth was first standardized in RFC6660 (which I co-authored). This 
had to be isolated from the e2e use of ECN by inclusion of a DSCP as well.


The relatively late addition of the fourth approach means that an 
attempt to mark using the SCE approach (10->01) is more likely to find 
that it gets reversed when the outer header is decapsulated, if the 
decapsulator hasn't been updated to the latest RFC that catered for this 
fourth transition (RFC6040, also co-authored by me).


L4S follows the original RFC3168 approach
SCE uses the fourth

So, SCE proposes to use /a/ correct approach, but it might not work.
Whereas L4S uses the original correct approach.

3a/ DualQ L4S AQMs
With the DualQ, the difference between the two queues is both in their 
ECN marking behaviour and in their forwarding/scheduling behaviour. 
However, whenever there's traffic in the classic queue the coupling 
between the AQMs overrides the network scheduler. The coupling is solely 
ECN behaviour not scheduling behaviour. So the primary difference 
between the queues is in their ECN-marking behaviour.


What do I mean by "the coupling overrides the network scheduler"? The 
network scheduler certainly does give priority to L4S packets whenever 
they arrive, but the coupling makes the L4S sources control how often 
packets arrive. It's tough to reason about, because we haven't had a 
mechanism like this before.


2b/ FQ L4S AQMs
If the AQM is implemented with per flow queues, the picture is clearer. 
The only difference between the queues is in the ECN marking behaviour 
of the different AQMs.




Bob


Regards
  Roland


--

Bob Briscoe   http://bobbriscoe.net/

___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Bless, Roland (TM)]

Hi,

Am 21.03.19 um 09:02 schrieb Bob Briscoe:
> Just to rapidly reply,
> 
> 
> On 21/03/2019 07:46, Jonathan Morton wrote:
>> The ECN field was never intended to be used as a classifier, except to
>> distinguish Not-ECT flows from ECT flows (which a middlebox does need
>> to know, to choose between mark and drop behaviours).  It was intended
>> to be used to convey congestion information from the network to the
>> receiver.  SCE adheres to that ideal.
> 
> Each PHB has a forwarding behaviour a DSCP re-marking behaviour and an
> ECN marking behaviour. The ECN field is the claissifer for the ECN
> marking behaviour.

That's exactly the reason, why using ECT(1) as classifier for L4S
behavior is not the right choice. L4S should use a DSCP for
classification, because it is actually defining a PHB.

Regards
 Roland
___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Sebastian Moeller]

> On Mar 21, 2019, at 07:04, Bob Briscoe  wrote:
> 
> [...]
> As regards the desire to use SCE instead of the L4S approach of using a 
> classifier, please answer all the reasons I gave for why that won't work, 
> which I sent in response to your draft some days ago.

Is there any work planned showing why the heuristic based on CE is not 
going to cause problems? Because as it stands ECT(1) might be a useful piece of 
information for a traffic classifier, but it certainly is not a reliable 
traffic category identifier (unless you are interested in the category: packets 
that promise to respond in the L4S-way if they should encounter congestion).

> The main one is incremental deployment: the source does not identify its 
> packets as distinct from others, so the source needs the network to use some 
> other identifier if it wants the network to put it in a queue with latency 
> that is isolated from packets not using the scheme. The only way I can see to 
> so this would be to use per-flow-queuing. I think that is an unstated 
> assumption of SCE.

You write a whole section on alternative labeling schemes in 
https://tools.ietf.org/html/draft-ietf-tsvwg-ecn-l4s-id-05#appendix-B that you 
rule out, but that is based on hand-waving over the fact that CE-marking 
destroys the neat L4S labeling by ECT(1) scheme in two ways (mis-classifiaction 
of by AQM and mis-interpretation by end-point). 
You write: "Given shortage of codepoints, both L4S and classic ECN sides of an 
AQM would have to use the same CE codepoint to indicate that a packet had 
experienced congestion.  If a packet that had already been marked CE in an 
upstream buffer arrived at a subsequent AQM, this AQM would then have to guess 
whether to classify CE packets as L4S or classic ECN.  Choosing the L4S 
treatment would be a safer choice, because then a few classic packets might 
arrive early, rather than a few L4S packets  arriving late;" but in all 
honestly this simply assumes the rate of CE-marked packets of non-L4S flows 
will be low, otherwise your four Ls will suffer.

Regarding the second ambiguity you write:
"A.1.4.  Fall back to Reno-friendly congestion control on classic ECN 
bottlenecks [side note on 
https://tools.ietf.org/html/draft-ietf-tsvwg-ecn-l4s-id-05#page-23 bottleneck 
appears in normal font instead of bold]
It would take time for endpoints to distinguish classic and L4S ECN marking.  
An increase in queuing delay or in delay variation would be a tell-tale sign, 
but it is not yet clear where a line would be drawn between the two behaviours. 
 It might be possible to cache what was learned about the path to help 
subsequent attempts to detect the type of marking."
This has issues that IMHO need empirical data instead of hand-waving. Competent 
AQM solutions will control traffic rates without introducing massively 
increasing delays which will make it harder to do a heuristic based on RTT or 
RTT variation, this is the same mis-design LEDBAT suffers from, making 
inportant decisions based on un-reliable information... And trying to cache the 
L4S-ness of the active AQM assumes a steady-state behaviour of the network, 
which will not work for all cases.

> 
> In contrast, L4S works with either per-flow queuing or dualQ, so it is more 
> appropriate for a wider spread of scenarios. Again, in that same unanswered 
> email, I described a way L4S can use per-flow queuing, and Greg has since 
> given pseudocode. There are other problems with doing the codepoints the SCE 
> way round - pls see that email.
> 
> There has been a general statement that the SCE way round is purer. However, 
> that concept is in the eye of the beholder. The SCE way round does not allow 
> the ECN field to be used as a classifier,

Nor does the proposed L4S interpretation, as elaborated above.

> so you don't get the benefit above about support for per-flow-queueing and 
> dual queue. You also don't get the benefit of being able to relax 
> resequencing in the network,

I am still failing to grasp how this is supposed to work, and I had a 
look at the RACK RFC already. IMHO the link technology people should think 
about how much slack they require and ask the RACK developers to add that as a 
minimum to the specification, assuming it is reasonable. Say, lower level ARQ 
is expected to take 5ms worst-case, than ask RACK to specify a minimum 
reordering window of 10ms. The current RACK draft with re-ordering window bound 
to be <= RTT will not give reliable re-odering allowance to the lower levels 
unless they measure the RTT for each flow independently, I am sure that that is 
not going to fly...


> because the network has no classifier to look at.

Same here CE-marked packets have no reliable label, and I assume that 
with L4S at steady-state and link capacity one can expect quite a number of 
CE-marked packets.
I begin to wonder whether there is a nomenclature mismatch here, and I have a 
definition of classific

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Bob Briscoe]

Just to rapidly reply,


On 21/03/2019 07:46, Jonathan Morton wrote:

The ECN field was never intended to be used as a classifier, except to 
distinguish Not-ECT flows from ECT flows (which a middlebox does need to know, 
to choose between mark and drop behaviours).  It was intended to be used to 
convey congestion information from the network to the receiver.  SCE adheres to 
that ideal.


Each PHB has a forwarding behaviour a DSCP re-marking behaviour and an 
ECN marking behaviour. The ECN field is the claissifer for the ECN 
marking behaviour.



Bob

--

Bob Briscoe   http://bobbriscoe.net/

___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Jonathan Morton]

> On 21 Mar, 2019, at 8:04 am, Bob Briscoe  wrote:

> Congestion controls are tricky to get stable in all situations. So it is 
> important to separate ideas and research from engineering of more mature 
> approaches that are ready for more widespread experimentation on the public 
> Internet. Our goal with L4S was to use proven algorithms, and put in place 
> mechanism to allow those algorithms to evolve.

I hope that from my example, you can see how to adapt a more flexible and 
"mature" version of DCTCP to use SCE.  You should be able to use the same 
algorithms that you've worked so hard on; only the signalling method changes, 
and the trigger for falling back to Classic ECN behaviour is explicit (a plain 
old CE mark).

As for "proven algorithms", it was conclusively proven that DCTCP was *not* 
compatible with Classic ECN middleboxes, and had only been proven to work in 
tightly controlled environments.  I am told that TCP Prague has a failsafe, but 
I do not yet understand how that failsafe works, and what I have been told 
sounds fragile.  I am honestly perplexed that no explanation of this is 
forthcoming.

SCE works transparently with every deployed and proven congestion control 
algorithm out there, which simply ignores the information SCE provides.  
Adaptations of some of those algorithms to incorporate SCE information seem to 
be straightforward to implement, especially since ns-3 now supports AccECN, so 
initial full-system experiments should be forthcoming quite soon.  We should 
even be able to rehabilitate DCTCP without resorting to failsafe workarounds - 
which *should* have you guys jumping for joy, in theory.

> As regards the desire to use SCE instead of the L4S approach of using a 
> classifier, please answer all the reasons I gave for why that won't work, 
> which I sent in response to your draft some days ago.

I'm afraid that must have got lost in the noise.  There *was* a lot of noise; 
it gave me a headache.

Regardless, I haven't seen any real claims that SCE won't work, except for some 
quibbles about RTT-fair convergence with single queues, which I subsequently 
found an elegant way to address.  We do have a bit of a publication bottleneck 
over here at the moment; limited manpower.

I have mainly seen claims that SCE isn't a one-for-one replacement for L4S 
using exactly the same mechanisms and infrastructure as L4S does.  Which is 
true, but unhelpful, because that would make SCE literally identical to L4S 
with no advantages of its own.  I'm willing to point out ways to implement L4S' 
goals using SCE; see below.

> The main one is incremental deployment: the source does not identify its 
> packets as distinct from others, so the source needs the network to use some 
> other identifier if it wants the network to put it in a queue with latency 
> that is isolated from packets not using the scheme. The only way I can see to 
> so this would be to use per-flow-queuing. I think that is an unstated 
> assumption of SCE.

Strictly minimising latency for the individual flow, in the face of competing 
non-SCE traffic sharing a single queue, is not a goal of SCE per se; I consider 
it an orthogonal problem which is better addressed by existing solutions.  
Coexisting with existing endpoints, existing traffic and existing middleboxes 
is paramount, and forms our main argument for incremental deployability.

Solutions already available include FQ and Diffserv.  I'll grant you that FQ is 
easier to implement at lowish speeds, where a cheap CPU can be loaded with 
flexible software to do the job.  You appear to be more focused on relatively 
high link capacities, as that is your main argument against FQ.  I'll just note 
in passing that good FQ can extract a lot of responsiveness from relatively 
low-capacity links.

Diffserv is widely deployed (in terms of hardware capabilities) and should be a 
natural fit for distinguishing classes of traffic from each other.  It is 
rarely used by applications because the networks tend to corrupt it in transit, 
and rarely make good use of the information into the bargain.  It strikes me 
that the cable industry may have more influence over that than I do.

> The SCE way round does not allow the ECN field to be used as a classifier…

The ECN field was never intended to be used as a classifier, except to 
distinguish Not-ECT flows from ECT flows (which a middlebox does need to know, 
to choose between mark and drop behaviours).  It was intended to be used to 
convey congestion information from the network to the receiver.  SCE adheres to 
that ideal.

There is a perfectly good and under-utilised 6-bit field for carrying 
classifier information, right there in the same byte as the ECN field.  You 
might want to ask the LE PHB guys for advice on making good use of it.

> You also don't get the benefit of being able to relax resequencing in the 
> network, because the network has no classifier to look at.

My position is that the network is already fre

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Bob Briscoe]

Jonathan,

On 20/03/2019 23:51, Jonathan Morton wrote:

On 21 Mar, 2019, at 1:29 am, Bob Briscoe  wrote:


But more importantly, the L4S usage couples the minimized latency use
case to any possibility of getting a high fidelity explicit congestion
signal, so the "maximize throughput" use case can't ever get it.

Eh? There's definitely a misunderstanding or a difference in terminology 
between us here. The whole point of using a congestion controller like DCTCP is 
so that flow rate can scale indefinitely with capacity. Van Jacobson actually 
noted that the original TCP was unscalable in a footnote to the tech report 
version of the SIGCOMM paper.

The high fidelity congestion signal of what we call scalable congestion 
controllers (like DCTCP) is inversely proportional to the window. So as window 
scales up, the congestion signal scales down, so that their product remains 
constant. That means the number of ECN marks per RTT is scale-invariant. So the 
control signal remains just as tight at any scale.

If you'll indulge me for a moment, I'd like to lay out a compromise scenario 
where a lot of L4S' stated goals are still met.

There is no dualQ.  There is an AQM at the bottleneck link, of unspecified 
type, which implements SCE.  Assume that it produces CE marks like a 
conventional AQM, and also produces SCE marks like an L4S AQM produces CE.

A sender implements DCTCP-SCE, which is essentially Paced NewReno modified to 
subtract half of all acked data that was SCE-marked from its cwnd.  (This is 
equivalent to the DCTCP algorithm with g=1 and an arbitrarily small measurement 
window, but acting on SCE instead of CE.)  Any SCE mark also kicks it out of 
slow-start.

The means by which SCE information gets back to the sender is left vague for 
now; it's an orthogonal problem with several viable solutions.

What is missing from this scenario, from L4S' point of view?  And why have I 
been able to describe it so succinctly?
My goal is also to tighten the EWMA parameter, g, in DCTCP to 1 (or 2). 
That is why we have recommended a queuing-time-based ramp AQM for the 
Low Latency queue, which so far works equivalently to the step with g 
set to its current default of 1/16. We have been doing experiments on 
this for some time. But it is important to assess each change one at a 
time.


Congestion controls are tricky to get stable in all situations. So it is 
important to separate ideas and research from engineering of more mature 
approaches that are ready for more widespread experimentation on the 
public Internet. Our goal with L4S was to use proven algorithms, and put 
in place mechanism to allow those algorithms to evolve.


As regards the desire to use SCE instead of the L4S approach of using a 
classifier, please answer all the reasons I gave for why that won't 
work, which I sent in response to your draft some days ago. The main one 
is incremental deployment: the source does not identify its packets as 
distinct from others, so the source needs the network to use some other 
identifier if it wants the network to put it in a queue with latency 
that is isolated from packets not using the scheme. The only way I can 
see to so this would be to use per-flow-queuing. I think that is an 
unstated assumption of SCE.


In contrast, L4S works with either per-flow queuing or dualQ, so it is 
more appropriate for a wider spread of scenarios. Again, in that same 
unanswered email, I described a way L4S can use per-flow queuing, and 
Greg has since given pseudocode. There are other problems with doing the 
codepoints the SCE way round - pls see that email.


There has been a general statement that the SCE way round is purer. 
However, that concept is in the eye of the beholder. The SCE way round 
does not allow the ECN field to be used as a classifier, so you don't 
get the benefit above about support for per-flow-queueing and dual 
queue. You also don't get the benefit of being able to relax 
resequencing in the network, because the network has no classifier to 
look at. For these, the SCE codepoint would need to be combined with a 
DSCP, and I assume you don't want to do that.


The L4S way round signifies an alternative meaning of the ECN field, 
which is exactly what it is. The problem of having to guess which type 
of packet a CE used to be has been roundly discussed at the IETF in TCPM 
and TSVWG WGs and it has been decided it is a non-problem if it is 
assumed to have been ECT(1) even if it was not - as written up in 
draft-ietf-ecn-l4s-id. And that discussion assumed TCP with 3DupACK 
reordering tolerance, not the more liberal use of RACK (or a RACK-like 
approach in other transports). With a RACK-like approach, it becomes 
even less of a problem.



Bob


  - Jonathan Morton



--

Bob Briscoe   http://bobbriscoe.net/

___
Bloat mailing list
Bloat@lists.bufferbloat.net
ht

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Jonathan Morton]

> On 21 Mar, 2019, at 1:29 am, Bob Briscoe  wrote:
> 
>> But more importantly, the L4S usage couples the minimized latency use
>> case to any possibility of getting a high fidelity explicit congestion
>> signal, so the "maximize throughput" use case can't ever get it.

> Eh? There's definitely a misunderstanding or a difference in terminology 
> between us here. The whole point of using a congestion controller like DCTCP 
> is so that flow rate can scale indefinitely with capacity. Van Jacobson 
> actually noted that the original TCP was unscalable in a footnote to the tech 
> report version of the SIGCOMM paper.
> 
> The high fidelity congestion signal of what we call scalable congestion 
> controllers (like DCTCP) is inversely proportional to the window. So as 
> window scales up, the congestion signal scales down, so that their product 
> remains constant. That means the number of ECN marks per RTT is 
> scale-invariant. So the control signal remains just as tight at any scale. 

If you'll indulge me for a moment, I'd like to lay out a compromise scenario 
where a lot of L4S' stated goals are still met.

There is no dualQ.  There is an AQM at the bottleneck link, of unspecified 
type, which implements SCE.  Assume that it produces CE marks like a 
conventional AQM, and also produces SCE marks like an L4S AQM produces CE.

A sender implements DCTCP-SCE, which is essentially Paced NewReno modified to 
subtract half of all acked data that was SCE-marked from its cwnd.  (This is 
equivalent to the DCTCP algorithm with g=1 and an arbitrarily small measurement 
window, but acting on SCE instead of CE.)  Any SCE mark also kicks it out of 
slow-start.

The means by which SCE information gets back to the sender is left vague for 
now; it's an orthogonal problem with several viable solutions.

What is missing from this scenario, from L4S' point of view?  And why have I 
been able to describe it so succinctly?

 - Jonathan Morton

___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Bob Briscoe]

se at once. 
Strictly it directly addresses all except minimize cost, but minimize 
cost can be built around the system. This was the subject of my PhD. I 
haven't described L4S in these terms, because most people are only 
interested in the latency. But this is the underlying reason for my 
obsession with ECN.


Frank Kelly predicted that queuing delay would be removed from the 
optimization as it was minimized. With L4S we've got very close to that.


ECN removes all congestion loss.

And the use of a inverse linear congestion controller gives the scalable 
throughput. I shall be touching on this in my talk for netdev tomorrow, 
but it's not really a subject for an implementation conference.


Minimize cost is something you do by combining the congestion signals 
across a network. So any AQM is part of that. And congestion controllers 
are the other part - they implicitly optimize cost, using the congestion 
signals as shadow prices. The square root in classic TCP distorts this, 
but DCTCP's inverse linear controller gives proportional fairness 
directly. Without a weight term in the congestion controller, there is 
not really an economic optimization, but that can be built onto a 
proportionally fair system and competition will gradually cause that to 
happen (or regulation as a proxy for competition). These are very long 
term processes though.



But more importantly, the L4S usage couples the minimized latency use

case to any possibility of getting a high fidelity explicit congestion

signal, so the "maximize throughput" use case can't ever get it.

Eh? There's definitely a misunderstanding or a difference in terminology 
between us here. The whole point of using a congestion controller like 
DCTCP is so that flow rate can scale indefinitely with capacity. Van 
Jacobson actually noted that the original TCP was unscalable in a 
footnote to the tech report version of the SIGCOMM paper.


The high fidelity congestion signal of what we call scalable congestion 
controllers (like DCTCP) is inversely proportional to the window. So as 
window scales up, the congestion signal scales down, so that their 
product remains constant. That means the number of ECN marks per RTT is 
scale-invariant. So the control signal remains just as tight at any scale.


Cheers



Bob


Regards,

Jake

PS:

If you get a chance, I'm still interested in seeing answers to my

questions about deployment mitigations on the tsvwg list:

https://mailarchive.ietf.org/arch/msg/tsvwg/TWOVpI-SvVsYVy0_U6K8R04eq3A

I'm not surprised if it slipped by unnoticed, there have been a lot of

emails on this.  But good answers to those questions would go a long way

toward easing my concerns about the urgency on this discussion.

PPS:

These issues are a bit sideways to my technical reasons for preferring

the SCE formulation of ECT(1), which have more to do with the confusing

semantics and proliferation of corner cases it creates for CE and ECE.

But I thought I’d ask about them since it seemed like maybe there was a

disconnect here.

*From: *Bob Briscoe 
*Date: *2019-03-18 at 18:07
*To: *"David P. Reed" , Vint Cerf 
*Cc: *tsvwg IETF list , bloat 

*Subject: *Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] 
Implementation and experimentation of TCP Prague/L4S hackaton at IETF104


David,

On 17/03/2019 18:07, David P. Reed wrote:

Vint -

BBR is the end-to-end control logic that adjusts the source rate
to match the share of the bolttleneck link it should use.

It depends on getting reliable current congestion information via
packet drops and/or ECN.

So the proposal by these guys (not the cable guys) is an attempt
to improve the quality of the congestion signal inserted by the
router with the bottleneck outbound link.

What do you mean 'not the cable guys'?
This thread was reasonably civil until this intervention.


THe cable guys are trying to get a "private" field in the IP
header for their own use.


There is nothing private about this codepoint, and there never has 
been. Here's some data points:


* The IP header codepoint in question (ECT(1) in the ECN field) was 
proposed for use as an alternative ECN behaviour in July 2105 in the 
IETF AQM WG and the IETF's transport area WG (which handles all ECN 
matters).
* A year later there followed a packed IETF BoF on the subject (after 
2 open Bar BoFs).
* Long discussion ensued on the merits of different IP header field 
combinations, on both these IETF lists, involving people active on 
this list (bloat), including Dave Taht, who is acknowledged for his 
contributions in the IETF draft.

* That was when it was decided that ECT(1) was most appropriate.
* The logic of the decision is written up in an appendix of 
draft-ietf-ecn-l4s-id.
* David Black, one of the co-chairs of the IETF's transport area WG 
and co-author of both the original ECN and Diffserv RFCs, 

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Jonathan Morton]

> On 20 Mar, 2019, at 11:48 pm, Greg White  wrote:
> 
> the dual queue mechanism is not the only way to support L4S and TCP Prague.  
> As we’ve mentioned a time or two, an fq_codel system can also support L4S and 
> TCP Prague.  I’m not aware that anyone has implemented it to test it out yet 
> (because so far most interest has been on dual-queue), but I think the 
> simplest version is:
>  
> At dequeue:
> If packet indicates ECT(1):
> If sojourn_time > L4S_threshold:
> Set CE*, and forward packet
> Else:
> Forward packet
> Else:
> Do all the normal CoDel stuff
>  
> In many cases, all of the packets in a single queue will either all be ECT(1) 
> or none of them will.  But, to handle VPNs (where a mix of ECT(1) and 
> non-ECT(1) packets could share a queue), I would think that including the ECN 
> field in the flow hash would keep those packets separate.   
>  
> *a more sophisticated approach would be to mark CE based on a ramp function 
> between a min_thresh and max_thresh, which could be implemented as a random 
> draw, or via a counting function

The above looks remarkably similar to our proof-of-concept for an SCE 
middlebox.  Essentially:

At dequeue:
Do all the normal Codel stuff
If packet (still) carries ECT(0):
If sojourn_time > SCE_threshold:
Set SCE
Forward packet

And yes, a ramp function between 0 and codel_target would work better.  The 
above gives us something to test against with minimum hassle.

 - Jonathan Morton

___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Greg White]

I responded to point 2 separately.  In response to point 1, the dual queue 
mechanism is not the only way to support L4S and TCP Prague.  As we’ve 
mentioned a time or two, an fq_codel system can also support L4S and TCP 
Prague.  I’m not aware that anyone has implemented it to test it out yet 
(because so far most interest has been on dual-queue), but I think the simplest 
version is:

At dequeue:
If packet indicates ECT(1):
If sojourn_time > L4S_threshold:
Set CE*, and forward packet
Else:
Forward packet
Else:
Do all the normal CoDel stuff

In many cases, all of the packets in a single queue will either all be ECT(1) 
or none of them will.  But, to handle VPNs (where a mix of ECT(1) and 
non-ECT(1) packets could share a queue), I would think that including the ECN 
field in the flow hash would keep those packets separate.

*a more sophisticated approach would be to mark CE based on a ramp function 
between a min_thresh and max_thresh, which could be implemented as a random 
draw, or via a counting function




From: Bloat  on behalf of "Holland, Jake" 

Date: Wednesday, March 20, 2019 at 1:06 PM
To: Bob Briscoe , "David P. Reed" , 
Vint Cerf 
Cc: tsvwg IETF list , bloat 
Subject: Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and 
experimentation of TCP Prague/L4S hackaton at IETF104

Hi Bob & Greg,

I agree there has been a reasonably open conversation about the L4S
work, and thanks for all your efforts to make it so.

However, I think there's 2 senses in which "private" might be fair that
I didn't see covered in your rebuttals (merging forks and including
Greg's rebuttal by reference from here:
https://lists.bufferbloat.net/pipermail/bloat/2019-March/009038.html )

Please note:
I don't consider these senses of "private" a disqualifying argument
against the use of L4S, though I do consider them costs that should be
taken into account (and of course opinions may differ here).

With that said, I wondered whether either of you have any responses that
speak to these points:


1. the L4S use of the ECT(1) codepoint can't be marked CE except by a
patent-protected AQM scheduler.

I understand that l4s-id suggests the possibility of an alternate
scheme.  However, comparing the MUSTs of the section 5 requirements
with the claims made by the patent seems to leave no room for an
alternate that would not infringe the patent claims, unless I'm missing
something?

https://tools.ietf.org/html/draft-ietf-tsvwg-ecn-l4s-id-06#section-5
https://patents.google.com/patent/US20170019343A1/en


2. the L4S use of the ECT(1) codepoint privileges the low latency use
case.

As Jonathan Morton pointed out, low latency is only one of several
known use cases that would be worthwhile to identify to an AQM
scheduler, which the L4S approach cannot be extended to handle:
- Minimize Latency
- Minimize Loss
- Minimize Cost
- Maximize Throughput

https://lists.bufferbloat.net/pipermail/ecn-sane/2019-March/66.html

I understand that "Minimize Cost" is perhaps covered by LEPHB, and that
operator tuning parameters for a dualq node can possibly allow for
tuning between minimizing loss and latency, as mentioned in section
4.1 of aqm-dualq, but since the signaling is bundled together, it can
only happen for one at a time, if I'm reading it right.

But more importantly, the L4S usage couples the minimized latency use
case to any possibility of getting a high fidelity explicit congestion
signal, so the "maximize throughput" use case can't ever get it.


Regards,
Jake

PS:
If you get a chance, I'm still interested in seeing answers to my
questions about deployment mitigations on the tsvwg list:
https://mailarchive.ietf.org/arch/msg/tsvwg/TWOVpI-SvVsYVy0_U6K8R04eq3A

I'm not surprised if it slipped by unnoticed, there have been a lot of
emails on this.  But good answers to those questions would go a long way
toward easing my concerns about the urgency on this discussion.

PPS:
These issues are a bit sideways to my technical reasons for preferring
the SCE formulation of ECT(1), which have more to do with the confusing
semantics and proliferation of corner cases it creates for CE and ECE.
But I thought I’d ask about them since it seemed like maybe there was a
disconnect here.


From: Bob Briscoe 
Date: 2019-03-18 at 18:07
To: "David P. Reed" , Vint Cerf 
Cc: tsvwg IETF list , bloat 
Subject: Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and 
experimentation of TCP Prague/L4S hackaton at IETF104

David,
On 17/03/2019 18:07, David P. Reed wrote:

Vint -



BBR is the end-to-end control logic that adjusts the source rate to match the 
share of the bolttleneck link it should use.



It depends on getting reliable current congestion information via packet drops 
and/or ECN.



So

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Holland, Jake]

On 2019-03-20, 12:58, "Stephen Hemminger"  wrote:
Has anyone done a detailed investigation for prior art?

I don't know, but for serendipitous reasons I recently came across this,
which may be interesting to those who would be interested in digging
further on that question:

http://www.statslab.cam.ac.uk/~frank/PAPERS/tac.pdf
"On packet marking at priority queues"
R. J. Gibbens and F. P. Kelly
IEEE Transactions on Automatic Control 47 (2002) 1016-1020.

I wasn't sure if it was technically prior art or not, and I'm offering
no opinion, but it sounded similar to me in some significant ways.

HTH.

-Jake

___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Stephen Hemminger]

On Wed, 20 Mar 2019 19:04:17 +
"Holland, Jake"  wrote:

> Hi Bob & Greg,
> 
> I agree there has been a reasonably open conversation about the L4S
> work, and thanks for all your efforts to make it so.
> 
> However, I think there's 2 senses in which "private" might be fair that
> I didn't see covered in your rebuttals (merging forks and including
> Greg's rebuttal by reference from here:
> https://lists.bufferbloat.net/pipermail/bloat/2019-March/009038.html )
> 
> Please note:
> I don't consider these senses of "private" a disqualifying argument
> against the use of L4S, though I do consider them costs that should be
> taken into account (and of course opinions may differ here).
> 
> With that said, I wondered whether either of you have any responses that
> speak to these points:
> 
> 
> 1. the L4S use of the ECT(1) codepoint can't be marked CE except by a
> patent-protected AQM scheduler.
> 
> I understand that l4s-id suggests the possibility of an alternate
> scheme.  However, comparing the MUSTs of the section 5 requirements
> with the claims made by the patent seems to leave no room for an
> alternate that would not infringe the patent claims, unless I'm missing
> something?
> 
> https://tools.ietf.org/html/draft-ietf-tsvwg-ecn-l4s-id-06#section-5
> https://patents.google.com/patent/US20170019343A1/en
> 

Has anyone done a detailed investigation for prior art?
The patent office does not do a good job of looking for prior art,
and the parties in the patent process are not motivated to look.

Other vendors often are not interested either because their own house of
cards built on patents of previous research might come falling down.

___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Holland, Jake]

Hi Bob & Greg,

I agree there has been a reasonably open conversation about the L4S
work, and thanks for all your efforts to make it so.

However, I think there's 2 senses in which "private" might be fair that
I didn't see covered in your rebuttals (merging forks and including
Greg's rebuttal by reference from here:
https://lists.bufferbloat.net/pipermail/bloat/2019-March/009038.html )

Please note:
I don't consider these senses of "private" a disqualifying argument
against the use of L4S, though I do consider them costs that should be
taken into account (and of course opinions may differ here).

With that said, I wondered whether either of you have any responses that
speak to these points:


1. the L4S use of the ECT(1) codepoint can't be marked CE except by a
patent-protected AQM scheduler.

I understand that l4s-id suggests the possibility of an alternate
scheme.  However, comparing the MUSTs of the section 5 requirements
with the claims made by the patent seems to leave no room for an
alternate that would not infringe the patent claims, unless I'm missing
something?

https://tools.ietf.org/html/draft-ietf-tsvwg-ecn-l4s-id-06#section-5
https://patents.google.com/patent/US20170019343A1/en


2. the L4S use of the ECT(1) codepoint privileges the low latency use
case.

As Jonathan Morton pointed out, low latency is only one of several
known use cases that would be worthwhile to identify to an AQM
scheduler, which the L4S approach cannot be extended to handle:
- Minimize Latency
- Minimize Loss
- Minimize Cost
- Maximize Throughput

https://lists.bufferbloat.net/pipermail/ecn-sane/2019-March/66.html

I understand that "Minimize Cost" is perhaps covered by LEPHB, and that
operator tuning parameters for a dualq node can possibly allow for
tuning between minimizing loss and latency, as mentioned in section
4.1 of aqm-dualq, but since the signaling is bundled together, it can
only happen for one at a time, if I'm reading it right.

But more importantly, the L4S usage couples the minimized latency use
case to any possibility of getting a high fidelity explicit congestion
signal, so the "maximize throughput" use case can't ever get it.


Regards,
Jake

PS:
If you get a chance, I'm still interested in seeing answers to my
questions about deployment mitigations on the tsvwg list:
https://mailarchive.ietf.org/arch/msg/tsvwg/TWOVpI-SvVsYVy0_U6K8R04eq3A

I'm not surprised if it slipped by unnoticed, there have been a lot of
emails on this.  But good answers to those questions would go a long way
toward easing my concerns about the urgency on this discussion.

PPS:
These issues are a bit sideways to my technical reasons for preferring
the SCE formulation of ECT(1), which have more to do with the confusing
semantics and proliferation of corner cases it creates for CE and ECE.
But I thought I’d ask about them since it seemed like maybe there was a
disconnect here.


From: Bob Briscoe 
Date: 2019-03-18 at 18:07
To: "David P. Reed" , Vint Cerf 
Cc: tsvwg IETF list , bloat 
Subject: Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and 
experimentation of TCP Prague/L4S hackaton at IETF104

David,
On 17/03/2019 18:07, David P. Reed wrote:

Vint -



BBR is the end-to-end control logic that adjusts the source rate to match the 
share of the bolttleneck link it should use.



It depends on getting reliable current congestion information via packet drops 
and/or ECN.



So the proposal by these guys (not the cable guys) is an attempt to improve the 
quality of the congestion signal inserted by the router with the bottleneck 
outbound link.
What do you mean 'not the cable guys'?
This thread was reasonably civil until this intervention.





THe cable guys are trying to get a "private" field in the IP header for their 
own use.

There is nothing private about this codepoint, and there never has been. Here's 
some data points:

* The IP header codepoint in question (ECT(1) in the ECN field) was proposed 
for use as an alternative ECN behaviour in July 2105 in the IETF AQM WG and the 
IETF's transport area WG (which handles all ECN matters).
* A year later there followed a packed IETF BoF on the subject (after 2 open 
Bar BoFs).
* Long discussion ensued on the merits of different IP header field 
combinations, on both these IETF lists, involving people active on this list 
(bloat), including Dave Taht, who is acknowledged for his contributions in the 
IETF draft.
* That was when it was decided that ECT(1) was most appropriate.
* The logic of the decision is written up in an appendix of 
draft-ietf-ecn-l4s-id.
* David Black, one of the co-chairs of the IETF's transport area WG and 
co-author of both the original ECN and Diffserv RFCs, wrote RFC8311 to lay out 
the process for reclaiming and reusing the necessary codepoints.
* This work and the process of freeing up codepoints

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Sebastian Moeller]

Dear All,


> On Mar 20, 2019, at 00:59, Dave Taht  wrote:
> 
> On Tue, Mar 19, 2019 at 5:44 AM Greg White  wrote:
>> [...]
>> But, L4S has been demonstrated in real equipment and in simulation, and 
>> leverages an existing congestion
> 
> Not under circumstances I can control. That's Not Science.
[...]

It would be great if the L4S project maybe could help kick-start independent 
testing, by creating an sharing two VMs one with the appropriate client side 
patches and one with a L4S aware AQM (probably curvy RED to avoid the patent 
issue, assuming the patent does not cover curvyRED). So that it is easier to 
"kick" the tiers in a way that tests what the L4S project considers compliant 
clients/AQM. Personally I am interested to see how robust and reliable the  
detection of non-L4S CE sources is and how well the L4S side of the AQM will 
tolerate CE-marked packets from non-L4S senders, or in other words how well the 
"isolation" works. And also how L4S endpoints will deal with SCE emitting AQMs 
on their path.
I admit that I have doubts that ECT(1), basically a single "constellation" of a 
2-bit bitfield can serve as a replacement for a single independent bit in a 
single-bit bit-field, that seems required for real isolation of flows of 
different ECN-response types.

Best Regards
Sebastian

___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Dave Taht]

senders and take advantage 
> of new sender features like reordering tolerance that can further drive down 
> latency in many common link technologies.

There's nothing in SCE that stops the exact same idea, except that the
behavior is chosen reciever side, not sender side.

> SCE will only work if the bottleneck link implements fq.  Some bottleneck 
> network gear will not be able to implement fq or will not implement it due to 
> its undesirable side effects (see section 6 of RFC 8290).

This is overbroad.

And I didn't write that section.

> L4S will work if the bottleneck link implements *either* fq or dual queue.

And this is not a demonstrated fact yet - *at all*  that needs to
yield to experiment and independent analysis.

That's a good start actually to writing up a good comparison document.
I'll save that and work on it, if
I get a talk slot for it.

>
>
> Beyond that, they are *very,very* similar.

Yes!

>
>
>
> But, L4S has been demonstrated in real equipment and in simulation, and 
> leverages an existing congestion

Not under circumstances I can control. That's Not Science.

>controller that is available in Linux and Windows (with some tweaks).

SCE improves on an existing AQM that the default in roughly 100% of
linux deployments now, all the way up to the new RHEL8 release, which
was the last distro to adopt it, so far as I know. It doesn't require
a specialized tcp to use, either.

>  SCE leverages a paragraph in a draft that describes a first guess about how 
> a congestion controller might work.

We were kind of in a hurry. 'round here, we usually write the code and
test it thoroughly for years, before submitting a draft.

We did part of this work originally with the original definition of
ce_threshold (back in 2013?) when we too had fiddled with DCTCP,
before pacing and BBR had arrived.

Running code is up on several servers now and we hope to have a live
test over the internet at ietf. Maybe tcp-fu will land. Hopefully
tcp-prague will land.

>
>
>
> L4S has defined a congestion feedback mechanism so that these congestion 
> signals can get back to the sender.  SCE offers that “we’ll propose something 
> later”.
>
>
>
> BBR currently does not listen to explicit congestion signals, but it could be 
> updated to do so (for either SCE or L4S).
>
>
>
> -Greg
>
>
>
>
>
> From: Bloat  on behalf of "David P. 
> Reed" 
> Date: Sunday, March 17, 2019 at 12:07 PM
> To: Vint Cerf 
> Cc: bloat , "ecn-s...@lists.bufferbloat.net" 
> 
> Subject: Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and 
> experimentation of TCP Prague/L4S hackaton at IETF104
>
>
>
> Vint -
>
>
>
> BBR is the end-to-end control logic that adjusts the source rate to match the 
> share of the bolttleneck link it should use.
>
>
>
> It depends on getting reliable current congestion information via packet 
> drops and/or ECN.
>
>
>
> So the proposal by these guys (not the cable guys) is an attempt to improve 
> the quality of the congestion signal inserted by the router with the 
> bottleneck outbound link.
>
>
>
> THe cable guys are trying to get a "private" field in the IP header for their 
> own use.
>
>
>
>
>
> -Original Message-
> From: "Vint Cerf" 
> Sent: Saturday, March 16, 2019 5:57pm
> To: "Holland, Jake" 
> Cc: "Mikael Abrahamsson" , "David P. Reed" 
> , "ecn-s...@lists.bufferbloat.net" 
> , "bloat" 
> Subject: Re: [Ecn-sane] [Bloat] [iccrg] Fwd: [tcpPrague] Implementation and 
> experimentation of TCP Prague/L4S hackaton at IETF104
>
> where does BBR fit into all this?
>
> v
>
>
>
> On Sat, Mar 16, 2019 at 5:39 PM Holland, Jake  wrote:
>
> On 2019-03-15, 11:37, "Mikael Abrahamsson"  wrote:
> L4S has a much better possibility of actually getting deployment into the
> wider Internet packet-moving equipment than anything being talked about
> here. Same with PIE as opposed to FQ_CODEL. I know it's might not be as
> good, but it fits better into actual silicon and it's being proposed by
> people who actually have better channels into the people setting hard
> requirements.
>
> I suggest you consider joining them instead of opposing them.
>
>
> Hi Mikael,
>
> I agree it makes sense that fq_anything has issues when you're talking
> about the OLT/CMTS/BNG/etc., and I believe it when you tell me PIE
> makes better sense there.
>
> But fq_x makes great sense and provides real value for the uplink in a
> home, small office, coffee shop, etc. (if you run the final rate limit
> on the home side of the 

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Sebastian Moeller]

> On Mar 19, 2019, at 05:44, Greg White  wrote:
> If I can boil this down for the people who are jumping into this without 
> reading the drafts:
>  
>   • Both L4S and SCE are attempting to provide congestion-controlled 
> senders with better congestion signals so that flows can achieve link 
> capacity without buffering delay. 
>   • Both are proposing to use ECT(1) as part of the mechanism, but to use 
> it in different ways.

SCE tries to encode information about the quantitative congestion state 
of the marking AQM into ECT(1), while L4S tries to use this as a general 
identifier of promised behavior as a receiver of CE marks, or rather as an 
indication that flows marked ECT(1) will not respond to CE marks as described 
in rfc3168. Which realistically means any non-L4S AQM needs to learn quickly to 
drop ECT(1) packets instead of marking them CE; that seems better controlled 
than waiting for a fall-back to rfc3168-compliant CE response due to a 
heuristic based on RTT variation.

>   • SCE’s usage of ECT(1) potentially allows an automatic fallback to 
> traditional Cubic behavior if the bottleneck link is a single-queue 
> classic-ECN AQM (do any of these exist?), whereas L4S will need to detect 
> such a condition via RTT measurement
>   • L4S’s usage of ECT(1) allows links to identify new senders and take 
> advantage of new sender features like reordering tolerance that can further 
> drive down latency in many common link technologies.

But L4S is incapable of _reliably_ classifying L4S flows/packets as 
CE-marked packets default to L4S-treatment. This indicates to me, that ECT(1) 
is not really suited as a reliable L4S identifier, what am I missing? 
This ambiguity leads to the question of the side-effects of this leaky 
classification: what about re-ordering of CE-marked packets? I hope that out of 
caution CE-marked packets will not be re-ordered as these are very much not 
guaranteed to employ RACK. (And tangentially, how is a link that desires more 
latitude for re-ordering going to deal with the RACK requirement to keep the 
re-ordering windows <= 1 RTT, given that RTTs over the internet differ from a 
few to dozens of ms. . Is there any study showing how RACK and re-ordering 
actually interact in real-life?) And how is it going to help a link in regards 
to re-ordering at all? It has been argued, that links do not differentiate 
flows at all, and assuming TCP traffic to coexist for a long time with 
(DC)TCP_Prague traffic, how can a link actually allow more re-ordering than 
currently tolerable without severely impacting the TCP flows? If it just 
transmits all ECT(1) packets in its queue things will be a bit better than now, 
but after the egress queue is emptied the link might still be stalled until the 
re-transmit of ECT(0) and CE marked packets is finished, no?


>   • SCE will only work if the bottleneck link implements fq.  Some 
> bottleneck network gear will not be able to implement fq or will not 
> implement it due to its undesirable side effects (see section 6 of RFC 8290).
>   • L4S will work if the bottleneck link implements *either* fq or dual 
> queue.

The proof ought to e in the pudding ;) is there data showing an working 
L4S fq-AQM?

>  
> Beyond that, they are *very,very* similar.  
>  
> But, L4S has been demonstrated in real equipment and in simulation, and 
> leverages an existing congestion controller that is available in Linux and 
> Windows (with some tweaks).  

As far as I can see the public git repository for TCP Prague is only a 
few days old so how could that be "available in Linux and Windows" right now, 
and one could similarly argue that it will only take a few tweaks to teach 
cubic how to deal with SCE.


So I have no pony in this race as I am outside of the field, but the L4S RFCs 
seem to promise more than they 



> SCE leverages a paragraph in a draft that describes a first guess about how a 
> congestion controller might work.
>  
> L4S has defined a congestion feedback mechanism so that these congestion 
> signals can get back to the sender.  SCE offers that “we’ll propose something 
> later”.
>  
> BBR currently does not listen to explicit congestion signals, but it could be 
> updated to do so (for either SCE or L4S).
>  
> -Greg
>  
>  
> From: Bloat  on behalf of "David P. 
> Reed" 
> Date: Sunday, March 17, 2019 at 12:07 PM
> To: Vint Cerf 
> Cc: bloat , "ecn-s...@lists.bufferbloat.net" 
> 
> Subject: Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and 
> experimentation of TCP Prague/L4S hackaton at IETF104
>  
> Vint -
>  
> BBR is the end-to-end control logic that adjusts the source rate to match the 
> share of the bolttleneck link it should use.
>  
> It depends on get

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Sebastian Moeller]

> On Mar 19, 2019, at 08:10, Jonathan Morton  wrote:
> 
>> On 19 Mar, 2019, at 7:52 am, Greg White  wrote:
>> 
>> L4S utilizes TCP Prague, which falls back to traditional congestion control 
>> if the bottleneck link doesn't provide isolation.  
> 
> You see, this is the part I find difficult to believe that it will operate 
> reliably.  For a start, I have seen no detailed public description of TCP 
> Prague, even though it has supposedly been in "open" development for so long. 
>  My most recent information is that it's essentially a slightly modified 
> DCTCP.
> 
> "  It would take time for endpoints to distinguish classic and L4S ECN
>   marking.  An increase in queuing delay or in delay variation would be
>   a tell-tale sign, but it is not yet clear where a line would be drawn
>   between the two behaviours.  "

IMHO this is caused by the fact that ECT(1) is simply not suitable as a 
classifier, as it will only reliably classify unmarked packets, anything marked 
CE will looses the destinction whether the flow considers itself L4S ready or 
not. Could anyone point me to the section in the L4S RFCs that discusses this, 
please?

https://datatracker.ietf.org/doc/draft-ietf-tsvwg-ecn-l4s-id/ has the following:
" Given shortage of codepoints, both L4S and classic ECN sides of
 an AQM would have to use the same CE codepoint to indicate that
 a packet had experienced congestion.  If a packet that had
 already been marked CE in an upstream buffer arrived at a
 subsequent AQM, this AQM would then have to guess whether to
 classify CE packets as L4S or classic ECN.  Choosing the L4S
 treatment would be a safer choice, because then a few classic
 packets might arrive early, rather than a few L4S packets
 arriving late;"

But how is the L4S queue actually maintaining its low latency if it just 
admitted an non-L4S flow? This _might_ work if CE marked packets are rare, but 
IMHO this confirms my assessment that ECT(1) is a terrible choice for a 
classifier bit.

Best Regards
Sebastian




> Internet history is littered with failed attempts at implementing 
> delay-sensitive TCPs.  I can immediately think of several reasons why delay 
> can and will vary for reasons other than the bottleneck not implementing an 
> isolated queue  (just ask the BBR devs).  The mere presence of a wifi link on 
> the path, even if it is never the bottleneck, would be a trivial and common 
> example.
> 
> So please explain (or point to good documentation) how TCP Prague robustly 
> avoids misbehaving in a standard ECN environment, as is presently deployed.
> 
> 
> SCE explicitly does not rely on specific changes in behaviour by endpoints.  
> It just provides a conduit of information from the network to the receiver, 
> in addition to standard ECN behaviour.  The receiver is free to ignore that 
> information, without harming the network, and will naturally behave normally 
> and safely when that information is absent.  We have a proof-of-concept 
> implementation (a trivial mod of sch_codel and sch_fq_codel) which 
> successfully passes this information across the Internet and works with (is 
> transparently ignored by) existing endpoints and middleboxes.
> 
> In short, SCE is incrementally deployable by design.
> 
> The broader system of feedback and modified congestion control, which I call 
> ELR (Explicit Load Regulation) as an umbrella term, offers benefits which, 
> yes, have not yet been proved - but which are straightforward in concept and 
> should be amenable to analysis.  It seems likely that some work from L4S can 
> be used in this context.
> 
> - Jonathan Morton
> 
> ___
> Bloat mailing list
> Bloat@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/bloat

___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Jonathan Morton]

> On 19 Mar, 2019, at 7:52 am, Greg White  wrote:
> 
> L4S utilizes TCP Prague, which falls back to traditional congestion control 
> if the bottleneck link doesn't provide isolation.  

You see, this is the part I find difficult to believe that it will operate 
reliably.  For a start, I have seen no detailed public description of TCP 
Prague, even though it has supposedly been in "open" development for so long.  
My most recent information is that it's essentially a slightly modified DCTCP.

"  It would take time for endpoints to distinguish classic and L4S ECN
   marking.  An increase in queuing delay or in delay variation would be
   a tell-tale sign, but it is not yet clear where a line would be drawn
   between the two behaviours.  "

Internet history is littered with failed attempts at implementing 
delay-sensitive TCPs.  I can immediately think of several reasons why delay can 
and will vary for reasons other than the bottleneck not implementing an 
isolated queue  (just ask the BBR devs).  The mere presence of a wifi link on 
the path, even if it is never the bottleneck, would be a trivial and common 
example.

So please explain (or point to good documentation) how TCP Prague robustly 
avoids misbehaving in a standard ECN environment, as is presently deployed.


SCE explicitly does not rely on specific changes in behaviour by endpoints.  It 
just provides a conduit of information from the network to the receiver, in 
addition to standard ECN behaviour.  The receiver is free to ignore that 
information, without harming the network, and will naturally behave normally 
and safely when that information is absent.  We have a proof-of-concept 
implementation (a trivial mod of sch_codel and sch_fq_codel) which successfully 
passes this information across the Internet and works with (is transparently 
ignored by) existing endpoints and middleboxes.

In short, SCE is incrementally deployable by design.

The broader system of feedback and modified congestion control, which I call 
ELR (Explicit Load Regulation) as an umbrella term, offers benefits which, yes, 
have not yet been proved - but which are straightforward in concept and should 
be amenable to analysis.  It seems likely that some work from L4S can be used 
in this context.

 - Jonathan Morton

___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Greg White]

On 3/18/19, 11:35 PM, "Jonathan Morton"  wrote:

From my standpoint, the major objection to L4S is that it is not 
incrementally deployable, because DCTCP starves conventional TCPs unless run 
through an isolated queue.  This is something we quickly realised when L4S was 
first announced.  It is simply not practical to require all middleboxes on the 
path to support L4S before L4S endpoints can safely be deployed, except in the 
isolated and very controlled environments where DCTCP was "proved".

[GW] But this isn't true!L4S utilizes TCP Prague, which falls back to 
traditional congestion control if the bottleneck link doesn't provide 
isolation.  

___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Jonathan Morton]

>   • SCE will only work if the bottleneck link implements fq.  Some 
> bottleneck network gear will not be able to implement fq or will not 
> implement it due to its undesirable side effects (see section 6 of RFC 8290).

> SCE leverages a paragraph in a draft that describes a first guess about how a 
> congestion controller might work.

We have an update in the works that should enable RTT-fair convergence with 
single-queue AQMs, whether they support SCE or not.  To do this using the 
simplest reasonable adjustments to existing congestion control algorithms, the 
setpoint is no longer fixed at 50% but varies with the cwnd of each flow.  And 
yes, we have worked out what those adjustments should look like in practice, 
but we haven't yet had time to run tests, or describe them in a nicely 
formatted I-D.

This update should also allow a very DCTCP-like congestion control algorithm to 
work correctly with SCE, as long as it acts conventionally with CE marks and 
only has the reduced response to SCE marks.  That's something that DCTCP itself 
currently does not do.

>   • SCE’s usage of ECT(1) potentially allows an automatic fallback to 
> traditional Cubic behavior if the bottleneck link is a single-queue 
> classic-ECN AQM (do any of these exist?), whereas L4S will need to detect 
> such a condition via RTT measurement

From my standpoint, the major objection to L4S is that it is not incrementally 
deployable, because DCTCP starves conventional TCPs unless run through an 
isolated queue.  This is something we quickly realised when L4S was first 
announced.  It is simply not practical to require all middleboxes on the path 
to support L4S before L4S endpoints can safely be deployed, except in the 
isolated and very controlled environments where DCTCP was "proved".

> I find it extremely disappointing that some people on this list are taking 
> such a negative attitude to the major development in their own field that 
> they seem not to have noticed since it first hit the limelight in 2015.

It is not at all true that we were unaware of L4S.  Rather, we quite reasonably 
believed it would never get traction in the IETF or in the Internet at large 
unless that problem was robustly solved - and we thought the obvious solution 
*was* to use ECT(1) as SCE does, and to fix Diffserv (so that it becomes e2e 
usable to some degree) or implement FQ if isolating low-latency-assured traffic 
is so important.

Incidentally, during that time we were implementing a good FQ system that is 
now in Linux and in commercial products. Granted, it isn't designed for the 
sort of high-capacity links where FQ is traditionally considered impractical.

> L4S has defined a congestion feedback mechanism so that these congestion 
> signals can get back to the sender.  SCE offers that “we’ll propose something 
> later”.

It should be straightforward to adjust AccECN so that it primarily carries SCE 
information instead of unnecessarily detailed CE information.  That's one 
obvious solution, which we hoped those already familiar with L4S would 
recognise without explicit prompting.

We have outlines of other feedback methods, still awaiting conversion to the 
proper document format, including one that doesn't require a new TCP option (I 
understand there are objections to such things, centred around paranoid 
firewalls) and which existing middleboxes and endpoints will transparently 
ignore (like the rest of SCE).  It uses the NS bit which was also freed up by 
the obsoleting of Nonce Sum.

The I-D presently available defines the SCE codepoint and very little else.  
This is due to separation of concerns.  Other I-Ds will define feedback 
mechanisms, detailed modifications to congestion control algorithms, and 
recommendations as to what AQMs should do.

Perhaps if we get a chance to work, instead of responding to manufactured 
outrage that we dare to propose something different, these extra documents 
might surface in time for discussion.

 - Jonathan Morton

___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Greg White]

That is ridiculous.

You clearly haven’t read the drafts, and so are speaking from a position of 
ignorance.  Please get informed before making statements like this.

There is *absolutely* nothing cable-specific or “private” about L4S.  It is 
being developed in an open forum, the IETF!!   Yes, the cable industry is 
adopting L4S – because we recognized its potential.  Others are too, and anyone 
can.  It is a totally open spec, and has been since the initial drafts came out 
of the RITE project.  The cable industry was not involved in RITE (in fact the 
technology was first demonstrated on DSL equipment), and we learned about L4S 
when the rest of the world did.  We decided to become early adopters.  Yes, we 
were quiet about the fact that we were planning on adopting it (until now).

If individuals drop out of participating in the IETF, they shouldn’t be upset 
if the IETF continues to make progress on developing Internet technology in 
their absence.  It seems pretty disingenuous for DT to form his own private 
working group to come up with an incompatible, and limited, alternative to the 
ongoing work in IETF, then spring it on the IETF and start this FUD war.

The craziest thing about this whole thread is that there is a heck of a lot in 
common between L4S and SCE.  More in common than different.  My initial belief 
was that we all had similar goals (eliminating buffering latency in the 
internet) and we’d be able to achieve a meeting of the minds on the best way to 
use ECT(1) to achieve it.  Now, I’m not so sure what DT’s motivation is.

If I can boil this down for the people who are jumping into this without 
reading the drafts:


  *   Both L4S and SCE are attempting to provide congestion-controlled senders 
with better congestion signals so that flows can achieve link capacity without 
buffering delay.
  *   Both are proposing to use ECT(1) as part of the mechanism, but to use it 
in different ways.
  *   SCE’s usage of ECT(1) potentially allows an automatic fallback to 
traditional Cubic behavior if the bottleneck link is a single-queue classic-ECN 
AQM (do any of these exist?), whereas L4S will need to detect such a condition 
via RTT measurement
  *   L4S’s usage of ECT(1) allows links to identify new senders and take 
advantage of new sender features like reordering tolerance that can further 
drive down latency in many common link technologies.
  *   SCE will only work if the bottleneck link implements fq.  Some bottleneck 
network gear will not be able to implement fq or will not implement it due to 
its undesirable side effects (see section 6 of RFC 8290).
  *   L4S will work if the bottleneck link implements *either* fq or dual queue.

Beyond that, they are *very,very* similar.

But, L4S has been demonstrated in real equipment and in simulation, and 
leverages an existing congestion controller that is available in Linux and 
Windows (with some tweaks).  SCE leverages a paragraph in a draft that 
describes a first guess about how a congestion controller might work.

L4S has defined a congestion feedback mechanism so that these congestion 
signals can get back to the sender.  SCE offers that “we’ll propose something 
later”.

BBR currently does not listen to explicit congestion signals, but it could be 
updated to do so (for either SCE or L4S).

-Greg


From: Bloat  on behalf of "David P. Reed" 

Date: Sunday, March 17, 2019 at 12:07 PM
To: Vint Cerf 
Cc: bloat , "ecn-s...@lists.bufferbloat.net" 

Subject: Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and 
experimentation of TCP Prague/L4S hackaton at IETF104


Vint -



BBR is the end-to-end control logic that adjusts the source rate to match the 
share of the bolttleneck link it should use.



It depends on getting reliable current congestion information via packet drops 
and/or ECN.



So the proposal by these guys (not the cable guys) is an attempt to improve the 
quality of the congestion signal inserted by the router with the bottleneck 
outbound link.



THe cable guys are trying to get a "private" field in the IP header for their 
own use.





-Original Message-
From: "Vint Cerf" 
Sent: Saturday, March 16, 2019 5:57pm
To: "Holland, Jake" 
Cc: "Mikael Abrahamsson" , "David P. Reed" 
, "ecn-s...@lists.bufferbloat.net" 
, "bloat" 
Subject: Re: [Ecn-sane] [Bloat] [iccrg] Fwd: [tcpPrague] Implementation and 
experimentation of TCP Prague/L4S hackaton at IETF104
where does BBR fit into all this?
v

On Sat, Mar 16, 2019 at 5:39 PM Holland, Jake 
mailto:jholl...@akamai.com>> wrote:
On 2019-03-15, 11:37, "Mikael Abrahamsson" 
mailto:swm...@swm.pp.se>> wrote:
L4S has a much better possibility of actually getting deployment into the
wider Internet packet-moving equipment than anything being talked about
here. Same with PIE as opposed to FQ_CODEL. I know it's might not be as
good, bu

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Dave Taht]

On Tue, Mar 19, 2019 at 2:07 AM Bob Briscoe  wrote:
>
> David,
>
> On 17/03/2019 18:07, David P. Reed wrote:
>
> Vint -
>
>
>
> BBR is the end-to-end control logic that adjusts the source rate to match the 
> share of the bolttleneck link it should use.
>
>
>
> It depends on getting reliable current congestion information via packet 
> drops and/or ECN.
>
>
>
> So the proposal by these guys (not the cable guys) is an attempt to improve 
> the quality of the congestion signal inserted by the router with the 
> bottleneck outbound link.
>
> What do you mean 'not the cable guys'?
> This thread was reasonably civil until this intervention.

I think the inventor of udp, and the e2e argument has good reason to
blow his top. I also think vint asked a reasonable question - has this
been tested vs bbr?

>
>
> THe cable guys are trying to get a "private" field in the IP header for their 
> own use.
>
>
> There is nothing private about this codepoint, and there never has been. 
> Here's some data points:

I hope 
https://github.com/dtaht/bufferbloat-rfcs/blob/master/sce/draft-morton-taht-tsvwg-sce.txt
gets comprehensively evaluated in comparison to the more private use
of ect_1 you are talking about,

> * The IP header codepoint in question (ECT(1) in the ECN field) was proposed 
> for use as an alternative ECN behaviour in July 2105 in the IETF AQM WG and 
> the IETF's transport area WG (which handles all ECN matters).

I called for the closure of that wg because it had become an endless
bikeshed. Since the aqm chairs refused to apply long honored
traditions of the ietf - like *running code*. I ended up forming a new
wg

https://www.bufferbloat.net/projects/ecn-sane/wiki/rules/

I like my rules a lot better than what ended up going on here. I do
hope more folk join my wg.

> * A year later there followed a packed IETF BoF on the subject (after 2 open 
> Bar BoFs).

From which many, including myself, ran screaming. I actually quit the
ietf last prague, and went off to implement real code in real
products. Shipping. I lost track of the numbers after they cracked
10m.

> * Long discussion ensued on the merits of different IP header field 
> combinations, on both these IETF lists, involving people active on this list 
> (bloat), including Dave Taht, who is acknowledged for his contributions in 
> the IETF draft.

I have called repeatedly that my name be removed from the l4s related
drafts, in private, and in public. I in no way endorsed this effort,
except as a (somewhat dubious) experiment.

> * That was when it was decided that ECT(1) was most appropriate.

Love the passive voice there. Everybody else had left the room.

> * The logic of the decision is written up in an appendix of 
> draft-ietf-ecn-l4s-id.

Yea, read that. Lousy logic. My notes from that period said - "prob
won't work in a container/vm/network namespace. design an experiment
to test it when running code arrives." Running code never did.

None of this stuff has been subjected to rigor we put the aqms through
in the aqm group. There's no public ns2, or ns3 models, no public
implementations at all... and y'all are proposing to write tcp prague
from scratch at a hackathon this weekend?

come on.

> * David Black, one of the co-chairs of the IETF's transport area WG and 
> co-author of both the original ECN and Diffserv RFCs, wrote RFC8311 to lay 
> out the process for reclaiming and reusing the necessary codepoints.

yea, that was good. used that for wireguard and now SCE.

> * This work and the process of freeing up codepoints has been very visible at 
> every IETF ever since, with multiple drafts to fix other aspects of the 
> protocols working their way through the IETF process in multiple WGs (tsvwg, 
> tcpm, trill, etc).

amazing levels of funding for that bit. Why on earth couldn't you hire
a decent hacker and get a version of tcpprague.c done 3 years ago?
You're Doing it at a hackathon next week? You know how much testing
even a simple change to tcp or and aqm goes through at google or
bufferbloat.net? Even with running code and widescale live testing?
I've been trying to analyze one tiny change to fq_codel on real
networks for close to a year now seems to work, so that's part of
the now SCE enabled repo here:

https://github.com/dtaht/fq_codel_fast


> * L4S has also been mentioned in IETF liaisons with the IEEE and 3GPP.
>
> Some history:
> * I had been researching the idea since 2012.
> * In fact my first presentation on it was scheduled directly after Van 
> Jacobson's first presentation of CoDel at the IETF in July 2012. VJ overran 
> by nearly 20mins leaving just 3 mins for my presentation.
> * Mirja Kuehlewind and I did early groundwork in 2013 and published a paper
> * Then I (working for BT) brought the work into the EU RITE project (Reducing 
> Internet Transport Latency) with Simula, Alcatel-Lucent, etc.
> * By 2015 the two main L4S proponents were Koen De Schepper from Alcatel 
> Lucent and myself (I had just switched from BT to Si

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Bob Briscoe]

David,

On 17/03/2019 18:07, David P. Reed wrote:


Vint -

BBR is the end-to-end control logic that adjusts the source rate to 
match the share of the bolttleneck link it should use.


It depends on getting reliable current congestion information via 
packet drops and/or ECN.


So the proposal by these guys (not the cable guys) is an attempt to 
improve the quality of the congestion signal inserted by the router 
with the bottleneck outbound link.



What do you mean 'not the cable guys'?
This thread was reasonably civil until this intervention.

THe cable guys are trying to get a "private" field in the IP header 
for their own use.




There is nothing private about this codepoint, and there never has been. 
Here's some data points:


* The IP header codepoint in question (ECT(1) in the ECN field) was 
proposed for use as an alternative ECN behaviour in July 2105 in the 
IETF AQM WG and the IETF's transport area WG (which handles all ECN 
matters).
* A year later there followed a packed IETF BoF on the subject (after 2 
open Bar BoFs).
* Long discussion ensued on the merits of different IP header field 
combinations, on both these IETF lists, involving people active on this 
list (bloat), including Dave Taht, who is acknowledged for his 
contributions in the IETF draft.

* That was when it was decided that ECT(1) was most appropriate.
* The logic of the decision is written up in an appendix of 
draft-ietf-ecn-l4s-id.
* David Black, one of the co-chairs of the IETF's transport area WG and 
co-author of both the original ECN and Diffserv RFCs, wrote RFC8311 to 
lay out the process for reclaiming and reusing the necessary codepoints.
* This work and the process of freeing up codepoints has been very 
visible at every IETF ever since, with multiple drafts to fix other 
aspects of the protocols working their way through the IETF process in 
multiple WGs (tsvwg, tcpm, trill, etc).

* L4S has also been mentioned in IETF liaisons with the IEEE and 3GPP.

Some history:
* I had been researching the idea since 2012.
* In fact my first presentation on it was scheduled directly after Van 
Jacobson's first presentation of CoDel at the IETF in July 2012. VJ 
overran by nearly 20mins leaving just 3 mins for my presentation.

* Mirja Kuehlewind and I did early groundwork in 2013 and published a paper
* Then I (working for BT) brought the work into the EU RITE project 
(Reducing Internet Transport Latency) with Simula, Alcatel-Lucent, etc.
* By 2015 the two main L4S proponents were Koen De Schepper from Alcatel 
Lucent and myself (I had just switched from BT to Simula), along with 
Olga Bondarenko (now Albisser), a PhD student at Simula who now works 
for Microsoft (on something else) and is still doing her PhD part-time 
with Simula

  o By that time, Al-Lu and Simula had a cool prototype running.
  o This was demonstrated publicly for the first time in the IETF AQM 
WG over DC+core+backhaul+DSL+home networks.

    https://riteproject.eu/dctth/#1511dispatchwg
* In May 2016, L4S was demonstrated at MultiMediaSystems'16 with /every/ 
packet from all the following simultaneous applications achieving ~1ms 
queuing delay or less over a 40Mb/s Internet access link (7ms base RTT):

  o cloud-rendered remote presence in a racing car within a VR headset
  o the interactive cloud-rendered video already linked above
  o an online gaming benchmark
  o HAS video streaming
  o a number of bulk file downloads
  o a heavy synthetic load of web browsing

L4S has never been access-technology-specific. Indeed the cable industry 
has been funding my work at the IETF to help encourage a wider L4S 
ecosystem. There is nothing private to the cable industry in this:
* Al-Lu used DSL as a use-case, but L4S was relevant to all the access 
technologies they supplied.

* BT was obviously interested in DSL,
* but BT's initial motivating use-case was to incrementally deploy the 
low queuing delay of DCTCP over BT's data centre interconnect networks.
* In Jul 2016 the open-source Linux repo for the Coupled AQM was 
published, with a fully working version to be used and abused.

   Now at: https://github.com/L4STeam/sch_dualpi2_upstream
* Of course, DCTCP was already open-sourced in Linux and FreeBSD, as 
well as available in Windows

* In Jul 2016, the main IETF BoF on L4S was held:
  o Ingemar Johansson from Ericsson was one of the proponents, focused 
on using L4S in LTE

  o along with Kevin Smith from Vodafone and
  o Praveen Balasubramanian from Microsoft (who maintains the Windows 
TCP stack, including DCTCP).
  o Ingemar has since written an open-source L4S variant of the SCReAM 
congestion controller for real-time media: 
https://github.com/EricssonResearch/scream/
  o Mirja Kuehlewind of ETHZ (and now Ericsson) implemented the 
necessary feedback in TCP https://github.com/mirjak/linux-accecn
* In summer 2017 CableLabs started work on Low Latency DOCSIS, and hired 
me later in the year to help develop and specify it, along with support 
for L

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Mikael Abrahamsson]

On Sun, 17 Mar 2019, Luca Muscariello wrote:


Fq_codel has an outstanding footprint in terms of deployment.


No, it doesn't.


A logical next step  to me seems to push chipcos to build fq_codel in
silicon.
It is totally feasible.


... and how do you plan to make that happen?

--
Mikael Abrahamssonemail: swm...@swm.pp.se
___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Toke Høiland-Jørgensen]

Luca Muscariello  writes:

> To me there is substantial difference from something like fq_codel or
> fq_pie where service differentiation is largely implicit
> and approches largely based on explicit marking.
>
> Approaches based on marking face technical and non technical challenges
> that have been largely mentioned in these lists.
>
> Fq_codel has a ton of litterature behind both theoretical and experimental
> and it is something very close to optimality, in terms of completion time
> and latency.
>
> Fq_codel also incentivizes the development of better congestion control as
> the reward is immediate. It also makes  Internet performance
> predictable.
>
> Once we know that, the logical approach would be to try to approximate that
> thing when the full mechanism is not possible because of a variety of
> limitations.
>
> This is the case in some DC switches that implement AFD+priority fair
> queueing at 40Gbps.
>
> Fq_codel has an outstanding footprint in terms of deployment.
> Iliad deployed SFQ in 2005 nation wide and Fq_codel as soon as it was
> available in France and is the second largest ISP.
> Iliad/Free  controls the development of both the home GW and the DSLAM.
> They have recently started to commercialize 10Gbps to the home using
> switched Ethernet.
> I’m very tempted to test it.
>
> Kudos to them for being able to prove it is possible as long as you control
> the development of your equipment.
>
> A logical next step  to me seems to push chipcos to build fq_codel in
> silicon.
> It is totally feasible.
>
> If on the other hand we say that we can achieve all fq_codel provides with
> current chipsets we’ll never create the incentives to make progress.

+100!

-Toke
___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Luca Muscariello]

To me there is substantial difference from something like fq_codel or
fq_pie where service differentiation is largely implicit
and approches largely based on explicit marking.

Approaches based on marking face technical and non technical challenges
that have been largely mentioned in these lists.

Fq_codel has a ton of litterature behind both theoretical and experimental
and it is something very close to optimality, in terms of completion time
and latency.

Fq_codel also incentivizes the development of better congestion control as
the reward is immediate. It also makes  Internet performance
predictable.

Once we know that, the logical approach would be to try to approximate that
thing when the full mechanism is not possible because of a variety of
limitations.

This is the case in some DC switches that implement AFD+priority fair
queueing at 40Gbps.

Fq_codel has an outstanding footprint in terms of deployment.
Iliad deployed SFQ in 2005 nation wide and Fq_codel as soon as it was
available in France and is the second largest ISP.
Iliad/Free  controls the development of both the home GW and the DSLAM.
They have recently started to commercialize 10Gbps to the home using
switched Ethernet.
I’m very tempted to test it.

Kudos to them for being able to prove it is possible as long as you control
the development of your equipment.

A logical next step  to me seems to push chipcos to build fq_codel in
silicon.
It is totally feasible.

If on the other hand we say that we can achieve all fq_codel provides with
current chipsets we’ll never create the incentives to make progress.

My2c
Luca

On Sun 17 Mar 2019 at 15:06, Mikael Abrahamsson  wrote:

> On Sat, 16 Mar 2019, Holland, Jake wrote:
>
> > Granted, it still remains to be seen whether SCE in practice can match
> > the results of L4S, and L4S was here first.  But it seems to me L4S comes
> > with some problems that have not yet been examined, and that are nicely
> > dodged by a SCE-based approach.
>
> I'm actually not that interested in an academic competition about what
> solution gives the ultimate "best" outcome in simulation or in a lab.
>
> I am interested in good enough solutions that are actually deployable and
> will get deployed, and doesn't have any pathological behaviour when it
> comes to legacy traffic.
>
> Right now the Internet is full of deep FIFOs and they're not going away,
> and they're not getting FQ_CODEL or CAKE.
>
> CAKE/FQ_CODEL is nice, but it's not being deployed at the typical
> congestion points we have in real life. These devices would have a much
> easier time getting PIE or even RED, if it was just implemented.
>
> --
> Mikael Abrahamssonemail: swm...@swm.pp.se
> ___
> Bloat mailing list
> Bloat@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/bloat
>
___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[David P. Reed]

Vint -
 
BBR is the end-to-end control logic that adjusts the source rate to match the 
share of the bolttleneck link it should use.
 
It depends on getting reliable current congestion information via packet drops 
and/or ECN.
 
So the proposal by these guys (not the cable guys) is an attempt to improve the 
quality of the congestion signal inserted by the router with the bottleneck 
outbound link.
 
THe cable guys are trying to get a "private" field in the IP header for their 
own use.
 
 
-Original Message-
From: "Vint Cerf" 
Sent: Saturday, March 16, 2019 5:57pm
To: "Holland, Jake" 
Cc: "Mikael Abrahamsson" , "David P. Reed" 
, "ecn-s...@lists.bufferbloat.net" 
, "bloat" 
Subject: Re: [Ecn-sane] [Bloat] [iccrg] Fwd: [tcpPrague] Implementation and 
experimentation of TCP Prague/L4S hackaton at IETF104



where does BBR fit into all this?
v


On Sat, Mar 16, 2019 at 5:39 PM Holland, Jake <[ jholl...@akamai.com ]( 
mailto:jholl...@akamai.com )> wrote:On 2019-03-15, 11:37, "Mikael Abrahamsson" 
<[ swm...@swm.pp.se ]( mailto:swm...@swm.pp.se )> wrote:
 L4S has a much better possibility of actually getting deployment into the 
 wider Internet packet-moving equipment than anything being talked about 
 here. Same with PIE as opposed to FQ_CODEL. I know it's might not be as 
 good, but it fits better into actual silicon and it's being proposed by 
 people who actually have better channels into the people setting hard 
 requirements.

 I suggest you consider joining them instead of opposing them.


 Hi Mikael,

 I agree it makes sense that fq_anything has issues when you're talking
 about the OLT/CMTS/BNG/etc., and I believe it when you tell me PIE
 makes better sense there.

 But fq_x makes great sense and provides real value for the uplink in a
 home, small office, coffee shop, etc. (if you run the final rate limit
 on the home side of the access link.)  I'm thinking maybe there's a
 disconnect here driven by the different use cases for where AQMs can go.

 The thing is, each of these is the most likely congestion point at
 different times, and it's worthwhile for each of them to be able to
 AQM (and mark packets) under congestion.

 One of the several things that bothers me with L4S is that I've seen
 precious little concern over interfering with the ability for another
 different AQM in-path to mark packets, and because it changes the
 semantics of CE, you can't have both working at the same time unless
 they both do L4S.

 SCE needs a lot of details filled in, but it's so much cleaner that it
 seems to me there's reasonably obvious answers to all (or almost all) of
 those detail questions, and because the semantics are so much cleaner,
 it's much easier to tell it's non-harmful.

 
 The point you raised in another thread about reordering is mostly
 well-taken, and a good counterpoint to the claim "non-harmful relative
 to L4S".

 To me it seems sad and dumb that switches ended up trying to make
 ordering guarantees at cost of switching performance, because if it's
 useful to put ordering in the switch, then it must be equally useful to
 put it in the receiver's NIC or OS.

 So why isn't it in all the receivers' NIC or OS (where it would render
 the switch's ordering efforts moot) instead of in all the switches?

 I'm guessing the answer is a competition trap for the switch vendors,
 plus "with ordering goes faster than without, when you benchmark the
 switch with typical load and current (non-RACK) receivers".

 If that's the case, it seems like the drive for a competitive advantage
 caused deployment of a packet ordering workaround in the wrong network
 location(s), out of a pure misalignment of incentives.

 RACK rates to fix that in the end, but a lot of damage is already done,
 and the L4S approach gives switches a flag that can double as proof that
 RACK is there on the receiver, so they can stop trying to order those
 packets.

 So point granted, I understand and agree there's a cost to abandoning
 that advantage.
 

 But as you also said so well in another thread, this is important.  ("The
 last unicorn", IIRC.)  How much does it matter if there's a feature that
 has value today, but only until RACK is widely deployed?  If you were
 convinced RACK would roll out everywhere within 3 years and SCE would
 produce better results than L4S over the following 15 years, would that
 change your mind?

 It would for me, and that's why I'd like to see SCE explored before
 making a call.  I think at its core, it provides the same thing L4S does
 (a high-fidelity explicit congestion signal for the sender), but with
 much cleaner semantics that can be incrementally added to congestion
 controls that people are already using.

 Granted, it still remains to be seen whether SCE in practice can match
 the results of L4S, and L4S was here first.  But it seems to me L4S comes
 with some problems that have not yet been examined, and that are nicely
 dodged by a SCE-based approach.

 If

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Dave Taht]

I helped land a more rfc - compliant version of pie into net-next late
last year.

ironically, we had one open bug re ecn -
https://github.com/gautamramk/FQ-PIE-for-Linux-Kernel/issues/2 -
docsis pie supported ecn not at all. I actually hold a good opinion of
pie - it is a good single queue aqm, more responsive to overload than
codel is. Which is why I worked on it to bring it up to spec finally.

Next up was polishing that version of fq-pie for linux inclusion. I
was unsure if it had the same codel-derived rate estimator as the bsd
one, because pie's original estimator fails with many queues. I always
wondered if that behavior would show up also in todays hw (64 on many
10GigE cards).

"Pie was added". Um, er, I (we) worked hard through 7 or so revisions
of the original quite crappy contractor written code, to make it
acceptable for mainline and for test. For free. In terms of billable
time, I was probably out $60k or more.

No "was" there. I know I shouldn't be resenting that phrasing, but I'd
like to obtain some credit for being fair and impartial. Also
thoroughly benchmarked it. and discarded it as I felt that a 5ms codel
target for good queue was better than a 20ms one for pie.

I'd offered multiple times to help fix up the dualpi code, but could
not look at it due to the frand patent, which I asked privately, be
removed.
___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Mikael Abrahamsson]

On Sun, 17 Mar 2019, Loganaden Velvindron wrote:

is there an open source implementation of PIE which is close to what is 
used by the DOCSIS modems ?


PIE was added to the Linux kernel in 3.14 according to 
https://www.linux.com/news/linux-314-release-no-pi-new-pie-fights-bufferbloat


--
Mikael Abrahamssonemail: swm...@swm.pp.se
___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Toke Høiland-Jørgensen]

On 17 March 2019 18:37:27 CET, Loganaden Velvindron  wrote:
>On Sun, Mar 17, 2019 at 6:06 PM Mikael Abrahamsson 
>wrote:
>>
>> On Sat, 16 Mar 2019, Holland, Jake wrote:
>>
>> > Granted, it still remains to be seen whether SCE in practice can
>match
>> > the results of L4S, and L4S was here first.  But it seems to me L4S
>comes
>> > with some problems that have not yet been examined, and that are
>nicely
>> > dodged by a SCE-based approach.
>>
>> I'm actually not that interested in an academic competition about
>what
>> solution gives the ultimate "best" outcome in simulation or in a lab.
>>
>> I am interested in good enough solutions that are actually deployable
>and
>> will get deployed, and doesn't have any pathological behaviour when
>it
>> comes to legacy traffic.
>>
>> Right now the Internet is full of deep FIFOs and they're not going
>away,
>> and they're not getting FQ_CODEL or CAKE.
>>
>> CAKE/FQ_CODEL is nice, but it's not being deployed at the typical
>> congestion points we have in real life. These devices would have a
>much
>> easier time getting PIE or even RED, if it was just implemented.
>>
>
>is there an open source implementation of PIE which is close to what
>is used by the DOCSIS modems ?

Yup. sch_pie in the Linux kernel. I believe Dave originally helped the Cisco 
people get it upstream...

There's even an out of tree fq_pie somewhere. Don't have the link handy.

-Toke

>
>> --
>> Mikael Abrahamssonemail: swm...@swm.pp.se
>> ___
>> Ecn-sane mailing list
>> ecn-s...@lists.bufferbloat.net
>> https://lists.bufferbloat.net/listinfo/ecn-sane
>___
>Ecn-sane mailing list
>ecn-s...@lists.bufferbloat.net
>https://lists.bufferbloat.net/listinfo/ecn-sane
___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Mikael Abrahamsson]

On Sat, 16 Mar 2019, Holland, Jake wrote:


Granted, it still remains to be seen whether SCE in practice can match
the results of L4S, and L4S was here first.  But it seems to me L4S comes
with some problems that have not yet been examined, and that are nicely
dodged by a SCE-based approach.


I'm actually not that interested in an academic competition about what 
solution gives the ultimate "best" outcome in simulation or in a lab.


I am interested in good enough solutions that are actually deployable and 
will get deployed, and doesn't have any pathological behaviour when it 
comes to legacy traffic.


Right now the Internet is full of deep FIFOs and they're not going away, 
and they're not getting FQ_CODEL or CAKE.


CAKE/FQ_CODEL is nice, but it's not being deployed at the typical 
congestion points we have in real life. These devices would have a much 
easier time getting PIE or even RED, if it was just implemented.


--
Mikael Abrahamssonemail: swm...@swm.pp.se
___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Sebastian Moeller]

> On Mar 16, 2019, at 22:38, Holland, Jake  wrote:
> 
> On 2019-03-15, 11:37, "Mikael Abrahamsson"  wrote:
>L4S has a much better possibility of actually getting deployment into the 
>wider Internet packet-moving equipment than anything being talked about 
>here. Same with PIE as opposed to FQ_CODEL. I know it's might not be as 
>good, but it fits better into actual silicon and it's being proposed by 
>people who actually have better channels into the people setting hard 
>requirements.
> 
>I suggest you consider joining them instead of opposing them.
> 
> 
> Hi Mikael,
> 
> I agree it makes sense that fq_anything has issues when you're talking
> about the OLT/CMTS/BNG/etc., and I believe it when you tell me PIE
> makes better sense there.

Except PIE is not mandatory there, DOCSIS3.1 made PIE mandatory on the 
CPE or customer modems, CMTS AQM was I believe recommended.

> 
> But fq_x makes great sense and provides real value for the uplink in a
> home, small office, coffee shop, etc. (if you run the final rate limit
> on the home side of the access link.)  I'm thinking maybe there's a
> disconnect here driven by the different use cases for where AQMs can go.
> 
> The thing is, each of these is the most likely congestion point at
> different times, and it's worthwhile for each of them to be able to
> AQM (and mark packets) under congestion.
> 
> One of the several things that bothers me with L4S is that I've seen
> precious little concern over interfering with the ability for another
> different AQM in-path to mark packets, and because it changes the
> semantics of CE, you can't have both working at the same time unless
> they both do L4S.

The relevant section from 
https://tools.ietf.org/html/draft-ietf-tsvwg-ecn-l4s-id-06:


"A.1.4.  Fall back to Reno-friendly congestion control on classic ECN

bottlenecks

   Description: A scalable congestion control needs to react to ECN
   marking from a non-L4S but ECN-capable bottleneck in a way that will
   coexist with a TCP Reno congestion control [RFC5681].

   Motivation: Similarly to the requirement in Appendix A.1.3, this
   requirement is a safety condition to ensure a scalable congestion
   control behaves properly when it builds a queue at a network
   bottleneck that has not been upgraded to support L4S.  On detecting
   classic ECN marking (see below), a scalable congestion control will
   need to fall back to classic congestion control behaviour.  If it
   does not comply with this requirement it could starve classic
   traffic.

   It would take time for endpoints to distinguish classic and L4S ECN
   marking.  An increase in queuing delay or in delay variation would be
   a tell-tale sign, but it is not yet clear where a line would be drawn
   between the two behaviours.  It might be possible to cache what was
   learned about the path to help subsequent attempts to detect the type
   of marking."


In short L4S has not seem to have solved this problem yet except for 
identifying it.
IMHO this is a clear reason not to to re-use ECT(1) outside of ECN signaling.


> 
> SCE needs a lot of details filled in, but it's so much cleaner that it
> seems to me there's reasonably obvious answers to all (or almost all) of
> those detail questions, and because the semantics are so much cleaner,
> it's much easier to tell it's non-harmful.

IMHO the beauty of the simple SCE proposal is that it simply supplies 
information a rational flow could/should react on purely by self interest, but 
ignoring it should do no harm, assuming the assumption holds that ECT(1) safely 
traverses the internet.

> 
> 
> The point you raised in another thread about reordering is mostly
> well-taken, and a good counterpoint to the claim "non-harmful relative
> to L4S".

Would this not be better handled by a dedicated signal instead of 
assuming all L4S traffic is re-ordering tolerant (which as seen from my vantage 
point runs counter L4S goal of ultra-low latency).


> 
> To me it seems sad and dumb that switches ended up trying to make
> ordering guarantees at cost of switching performance, because if it's
> useful to put ordering in the switch, then it must be equally useful to
> put it in the receiver's NIC or OS.

The issue I see, is that re-ordering with fast ARQ cycles on a fast 
link will be faster than pushing the un-ordered packets over the bottleneck 
access link, as in the case of data stretching over multiple packets the user 
might need them all before the data can be actually used.

> 
> So why isn't it in all the receivers' NIC or OS (where it would render
> the switch's ordering efforts moot) instead of in all the switches?
> 
> I'm guessing the answer is a competition trap for the switch vendors,
> plus "with ordering goes faster than without, when you benchmark the
> switch with typical load and current (non-RACK) receivers".
> 
> If that's the case, it seems like the drive for a competitive advant

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Holland, Jake]

Yeah, great question.

I don't want to answer for the L4S guys, I don't have a good feel for
what they might think.  But it does concern me that there seems to be at
least one tuning parameter that was picked for Reno, which I think I
mentioned on the tsvwg list:
https://tools.ietf.org/html/draft-ietf-tsvwg-aqm-dualq-coupled-08#section-2.1

For SCE, I would assume they'll want to make use of it at some point,
and so they'll have to write a draft for how BBR will handle it.

I think there's an open question of what exactly the rate of SCE
markings would look like for a SCE-capable AQM, and presumably this also
needs to be nailed down before it can be useful.  My initial instinct is
a probabilistic SCE setting based on current queue length, either when
forwarded or when enqueued, but I think this will take some more
thought, and I'm not sure that's best.

But whatever the most informative schedule we can figure out is, if that
info can get back to sender, it can essentially do whatever it thinks is
right, according to the cc it’s running, is my understanding.

-Jake


From: Vint Cerf 
Date: 2019-03-16 at 14:57
To: "Holland, Jake" 
Cc: Mikael Abrahamsson , "David P. Reed" 
, "ecn-s...@lists.bufferbloat.net" 
, bloat 
Subject: Re: [Ecn-sane] [Bloat] [iccrg] Fwd: [tcpPrague] Implementation and 
experimentation of TCP Prague/L4S hackaton at IETF104

where does BBR fit into all this?

v


On Sat, Mar 16, 2019 at 5:39 PM Holland, Jake 
mailto:jholl...@akamai.com>> wrote:
On 2019-03-15, 11:37, "Mikael Abrahamsson" 
mailto:swm...@swm.pp.se>> wrote:
L4S has a much better possibility of actually getting deployment into the
wider Internet packet-moving equipment than anything being talked about
here. Same with PIE as opposed to FQ_CODEL. I know it's might not be as
good, but it fits better into actual silicon and it's being proposed by
people who actually have better channels into the people setting hard
requirements.

I suggest you consider joining them instead of opposing them.


Hi Mikael,

I agree it makes sense that fq_anything has issues when you're talking
about the OLT/CMTS/BNG/etc., and I believe it when you tell me PIE
makes better sense there.

But fq_x makes great sense and provides real value for the uplink in a
home, small office, coffee shop, etc. (if you run the final rate limit
on the home side of the access link.)  I'm thinking maybe there's a
disconnect here driven by the different use cases for where AQMs can go.

The thing is, each of these is the most likely congestion point at
different times, and it's worthwhile for each of them to be able to
AQM (and mark packets) under congestion.

One of the several things that bothers me with L4S is that I've seen
precious little concern over interfering with the ability for another
different AQM in-path to mark packets, and because it changes the
semantics of CE, you can't have both working at the same time unless
they both do L4S.

SCE needs a lot of details filled in, but it's so much cleaner that it
seems to me there's reasonably obvious answers to all (or almost all) of
those detail questions, and because the semantics are so much cleaner,
it's much easier to tell it's non-harmful.


The point you raised in another thread about reordering is mostly
well-taken, and a good counterpoint to the claim "non-harmful relative
to L4S".

To me it seems sad and dumb that switches ended up trying to make
ordering guarantees at cost of switching performance, because if it's
useful to put ordering in the switch, then it must be equally useful to
put it in the receiver's NIC or OS.

So why isn't it in all the receivers' NIC or OS (where it would render
the switch's ordering efforts moot) instead of in all the switches?

I'm guessing the answer is a competition trap for the switch vendors,
plus "with ordering goes faster than without, when you benchmark the
switch with typical load and current (non-RACK) receivers".

If that's the case, it seems like the drive for a competitive advantage
caused deployment of a packet ordering workaround in the wrong network
location(s), out of a pure misalignment of incentives.

RACK rates to fix that in the end, but a lot of damage is already done,
and the L4S approach gives switches a flag that can double as proof that
RACK is there on the receiver, so they can stop trying to order those
packets.

So point granted, I understand and agree there's a cost to abandoning
that advantage.


But as you also said so well in another thread, this is important.  ("The
last unicorn", IIRC.)  How much does it matter if there's a feature that
has value today, but only until RACK is widely deployed?  If you were
convinced RACK would roll out everywhere within 3 years and SCE would
produce better results than L4S over the following 15 years, would that
change your mind?

It would for me, and that's why I'd like to see SCE explored before
making a call.  I think at its core, it provides the same thing L4S does
(a high-fidel

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Jonathan Morton]

> On 16 Mar, 2019, at 11:38 pm, Holland, Jake  wrote:
> 
> SCE needs a lot of details filled in, but it's so much cleaner that it
> seems to me there's reasonably obvious answers to all (or almost all) of
> those detail questions, and because the semantics are so much cleaner,
> it's much easier to tell it's non-harmful.

And we're actively working on filling in those details.  They haven't yet made 
it as far as an I-D, partly because we'd like to do at least some preliminary 
testing first.

> where does BBR fit into all this?

The present version of BBR, which I've actually seen, ignores ECN completely.  
I'm told that the new version which I haven't yet seen, does *something* with 
ECN, but I'm unclear on what.  I think it'll be up to the BBR developers to 
decide how to incorporate SCE information, using conventional TCPs as a guide 
as to what is reasonable - or to ignore it, which is also a valid design choice.

 - Jonathan Morton

___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Dave Taht]

Dear Vint:

BBR, along with all "non ect_1 sending L4S compatable" transports,
gets relegated to the dualpi  "Classic" queue.

https://datatracker.ietf.org/doc/draft-ietf-tsvwg-aqm-dualq-coupled/

On Sat, Mar 16, 2019 at 2:57 PM Vint Cerf  wrote:
>
> where does BBR fit into all this?
>
> v
>
>
> On Sat, Mar 16, 2019 at 5:39 PM Holland, Jake  wrote:
>>
>> On 2019-03-15, 11:37, "Mikael Abrahamsson"  wrote:
>> L4S has a much better possibility of actually getting deployment into the
>> wider Internet packet-moving equipment than anything being talked about
>> here. Same with PIE as opposed to FQ_CODEL. I know it's might not be as
>> good, but it fits better into actual silicon and it's being proposed by
>> people who actually have better channels into the people setting hard
>> requirements.
>>
>> I suggest you consider joining them instead of opposing them.
>>
>>
>> Hi Mikael,
>>
>> I agree it makes sense that fq_anything has issues when you're talking
>> about the OLT/CMTS/BNG/etc., and I believe it when you tell me PIE
>> makes better sense there.
>>
>> But fq_x makes great sense and provides real value for the uplink in a
>> home, small office, coffee shop, etc. (if you run the final rate limit
>> on the home side of the access link.)  I'm thinking maybe there's a
>> disconnect here driven by the different use cases for where AQMs can go.
>>
>> The thing is, each of these is the most likely congestion point at
>> different times, and it's worthwhile for each of them to be able to
>> AQM (and mark packets) under congestion.
>>
>> One of the several things that bothers me with L4S is that I've seen
>> precious little concern over interfering with the ability for another
>> different AQM in-path to mark packets, and because it changes the
>> semantics of CE, you can't have both working at the same time unless
>> they both do L4S.
>>
>> SCE needs a lot of details filled in, but it's so much cleaner that it
>> seems to me there's reasonably obvious answers to all (or almost all) of
>> those detail questions, and because the semantics are so much cleaner,
>> it's much easier to tell it's non-harmful.
>>
>> 
>> The point you raised in another thread about reordering is mostly
>> well-taken, and a good counterpoint to the claim "non-harmful relative
>> to L4S".
>>
>> To me it seems sad and dumb that switches ended up trying to make
>> ordering guarantees at cost of switching performance, because if it's
>> useful to put ordering in the switch, then it must be equally useful to
>> put it in the receiver's NIC or OS.
>>
>> So why isn't it in all the receivers' NIC or OS (where it would render
>> the switch's ordering efforts moot) instead of in all the switches?
>>
>> I'm guessing the answer is a competition trap for the switch vendors,
>> plus "with ordering goes faster than without, when you benchmark the
>> switch with typical load and current (non-RACK) receivers".
>>
>> If that's the case, it seems like the drive for a competitive advantage
>> caused deployment of a packet ordering workaround in the wrong network
>> location(s), out of a pure misalignment of incentives.
>>
>> RACK rates to fix that in the end, but a lot of damage is already done,
>> and the L4S approach gives switches a flag that can double as proof that
>> RACK is there on the receiver, so they can stop trying to order those
>> packets.
>>
>> So point granted, I understand and agree there's a cost to abandoning
>> that advantage.
>> 
>>
>> But as you also said so well in another thread, this is important.  ("The
>> last unicorn", IIRC.)  How much does it matter if there's a feature that
>> has value today, but only until RACK is widely deployed?  If you were
>> convinced RACK would roll out everywhere within 3 years and SCE would
>> produce better results than L4S over the following 15 years, would that
>> change your mind?
>>
>> It would for me, and that's why I'd like to see SCE explored before
>> making a call.  I think at its core, it provides the same thing L4S does
>> (a high-fidelity explicit congestion signal for the sender), but with
>> much cleaner semantics that can be incrementally added to congestion
>> controls that people are already using.
>>
>> Granted, it still remains to be seen whether SCE in practice can match
>> the results of L4S, and L4S was here first.  But it seems to me L4S comes
>> with some problems that have not yet been examined, and that are nicely
>> dodged by a SCE-based approach.
>>
>> If L4S really is as good as they seem to think, I could imagine getting
>> behind it, but I don't think that's proven yet.  I'm not certain, but
>> all the comparative analyses I remember seeing have been from more or
>> less the same team, and I'm not convinced they don't have some
>> misaligned incentives of their own.
>>
>> I understand a lot of work has gone into L4S, but this move to jump it
>> from interesting experiment to de-facto standard without a more critical
>> review that digs deeper i

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Holland, Jake]

On 2019-03-15, 11:37, "Mikael Abrahamsson"  wrote:
L4S has a much better possibility of actually getting deployment into the 
wider Internet packet-moving equipment than anything being talked about 
here. Same with PIE as opposed to FQ_CODEL. I know it's might not be as 
good, but it fits better into actual silicon and it's being proposed by 
people who actually have better channels into the people setting hard 
requirements.

I suggest you consider joining them instead of opposing them.


Hi Mikael,

I agree it makes sense that fq_anything has issues when you're talking
about the OLT/CMTS/BNG/etc., and I believe it when you tell me PIE
makes better sense there.

But fq_x makes great sense and provides real value for the uplink in a
home, small office, coffee shop, etc. (if you run the final rate limit
on the home side of the access link.)  I'm thinking maybe there's a
disconnect here driven by the different use cases for where AQMs can go.

The thing is, each of these is the most likely congestion point at
different times, and it's worthwhile for each of them to be able to
AQM (and mark packets) under congestion.

One of the several things that bothers me with L4S is that I've seen
precious little concern over interfering with the ability for another
different AQM in-path to mark packets, and because it changes the
semantics of CE, you can't have both working at the same time unless
they both do L4S.

SCE needs a lot of details filled in, but it's so much cleaner that it
seems to me there's reasonably obvious answers to all (or almost all) of
those detail questions, and because the semantics are so much cleaner,
it's much easier to tell it's non-harmful.


The point you raised in another thread about reordering is mostly
well-taken, and a good counterpoint to the claim "non-harmful relative
to L4S".

To me it seems sad and dumb that switches ended up trying to make
ordering guarantees at cost of switching performance, because if it's
useful to put ordering in the switch, then it must be equally useful to
put it in the receiver's NIC or OS.

So why isn't it in all the receivers' NIC or OS (where it would render
the switch's ordering efforts moot) instead of in all the switches?

I'm guessing the answer is a competition trap for the switch vendors,
plus "with ordering goes faster than without, when you benchmark the
switch with typical load and current (non-RACK) receivers".

If that's the case, it seems like the drive for a competitive advantage
caused deployment of a packet ordering workaround in the wrong network
location(s), out of a pure misalignment of incentives.

RACK rates to fix that in the end, but a lot of damage is already done,
and the L4S approach gives switches a flag that can double as proof that
RACK is there on the receiver, so they can stop trying to order those
packets.

So point granted, I understand and agree there's a cost to abandoning
that advantage.


But as you also said so well in another thread, this is important.  ("The
last unicorn", IIRC.)  How much does it matter if there's a feature that
has value today, but only until RACK is widely deployed?  If you were
convinced RACK would roll out everywhere within 3 years and SCE would
produce better results than L4S over the following 15 years, would that
change your mind?

It would for me, and that's why I'd like to see SCE explored before
making a call.  I think at its core, it provides the same thing L4S does
(a high-fidelity explicit congestion signal for the sender), but with
much cleaner semantics that can be incrementally added to congestion
controls that people are already using.

Granted, it still remains to be seen whether SCE in practice can match
the results of L4S, and L4S was here first.  But it seems to me L4S comes
with some problems that have not yet been examined, and that are nicely
dodged by a SCE-based approach.

If L4S really is as good as they seem to think, I could imagine getting
behind it, but I don't think that's proven yet.  I'm not certain, but
all the comparative analyses I remember seeing have been from more or
less the same team, and I'm not convinced they don't have some
misaligned incentives of their own.

I understand a lot of work has gone into L4S, but this move to jump it
from interesting experiment to de-facto standard without a more critical
review that digs deeper into some of the potential deployment problems
has me concerned.

If it really does turn out to be good enough to be permanent, I'm not
opposed to it, but I'm just not convinced that it's non-harmful, and my
default position is that the cleaner solution is going to be better in
the long run, if they can do the same job.

It's not that I want it to be a fight, but I do want to end up with the
best solution we can get.  We only have the one internet.

Just my 2c.  

-Jake


___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listin

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Michael Richardson]

The DSCP code points were also supposed to be local to the AS.
They are *supposed* to be recoded at the edges.

What is missing the diffedge protocol... MS actually implemented this in Win2K.
We have no way to signal the ToS we desire with the ISP, and so they are
forced to guess, and they do it wrong.

I actually see this as the critical NetNeutrality issue.  We *DO*
want multiple classes of services, but we want to be in charge of
what traffic goes into each class, not the ISP.

--
]   Never tell me the odds! | ipv6 mesh networks [
]   Michael Richardson, Sandelman Software Works|IoT architect   [
] m...@sandelman.ca  http://www.sandelman.ca/|   ruby on rails[



signature.asc
Description: PGP signature
___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Jonathan Foulkes]

Thanks Nils, I’d seen that PolicyPlus tool before, and while a nice stand-alone 
option for any Windows version, it is still a tool for admins. I think we need 
something regular technical users can use to do things like ‘Make DropBox 
traffic land in the bulk tin’.

Also, there is a good thread about other mechanisms to route traffic into the 
bulk tin using rules in the router here: 
https://github.com/dtaht/sch_cake/issues/97  Including a documented approach to 
setting windows DSCP settings via PowerShell.

Cheers,

Jonathan Foulkes

> On Mar 16, 2019, at 6:23 AM, Nils Andreas Svee  wrote:
> 
> You can use group policies on standalone clients, however the Local Group 
> Policy Editor is not available on the Windows Home editions (neither is 
> domain membership AFAIK), and so many people resort to applying the changes 
> they require directly to the registry.
> 
> There are also tools like Policy Plus that allows you to edit local GPOs 
> regardless of which edition of Windows you are running , but I haven't tried 
> it myself: https://github.com/Fleex255/PolicyPlus
> 
> Best Regards
> Nils
> ___
> Bloat mailing list
> Bloat@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/bloat

___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Nils Andreas Svee]

You can use group policies on standalone clients, however the Local Group 
Policy Editor is not available on the Windows Home editions (neither is domain 
membership AFAIK), and so many people resort to applying the changes they 
require directly to the registry.

There are also tools like Policy Plus that allows you to edit local GPOs 
regardless of which edition of Windows you are running , but I haven't tried it 
myself: https://github.com/Fleex255/PolicyPlus

Best Regards
Nils
___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Sebastian Moeller]

> On Mar 16, 2019, at 10:42, Michael Welzl  wrote:
> 
> Good question!  …. on Windows in particular, I’d really like to know this too.

Well, as far as I can tell it is the group policy editor that is the 
tool to assign DSCPs to applications, IMHO that is exactly the right place, 
somewhere where the administrator/enduser can set her desired policy 
(personally I am fine with applications also using sensible defaults, as long 
as the user can override them all is well). The catch seems to be that group 
policies require a domain controller and are hence not available on stand-alond 
windows home installations. Anybody with deep contacts to microsoft here, that 
could try to get an sub-official standpoint from MS on the issue of opening the 
group policy editor up for everybody (at least the dscp marking part)?

Best Regards
Sebastian


> 
> The WebRTC Javascript API allows one to influence the DSCP, i.e. browsers 
> normally can do that. Whether that’s true for all OSes, I don’t know.
> 
> Cheers,
> Michael
> 
> 
> 
>> On Mar 16, 2019, at 12:45 AM, David P. Reed  wrote:
>> 
>> How many applications used by normal users have "admin" privileges? The 
>> Browser? Email? FTP?
>>  
>>  
>> -Original Message-
>> From: "Dave Taht" 
>> Sent: Friday, March 15, 2019 4:31pm
>> To: "Jonathan Foulkes" 
>> Cc: ecn-s...@lists.bufferbloat.net, "bloat" 
>> Subject: Re: [Ecn-sane] [Bloat] [iccrg] Fwd: [tcpPrague] Implementation and 
>> experimentation of TCP Prague/L4S hackaton at IETF104
>> 
>> On Fri, Mar 15, 2019 at 1:28 PM Jonathan Foulkes  
>> wrote:
>> >
>> > All this discussion of DSCP marking brings to mind what happened on the 
>> > Windows platform, where the OS had to suppress ALL DSCP marks, as app 
>> > authors were trying to game the system.
>> > And even if not trying to ‘game’ it, they have non-obvious reasons why 
>> > they don’t mark traffic how one would expect. Example:
>> >
>> > I know an engineer who works at a cloud-storage solution company, and I 
>> > asked why a long-standing customer request for DSCP marking (as bulk) was 
>> > not implemented. His answer was they’d never do that, as that would impact 
>> > benchmarks against their competitors for which service syncs faster. 
>> >
>> > Which brings me to a question: Is anyone aware of an easy to use Windows 
>> > app that will allow the user to select an application and tell the OS to 
>> > mark the traffic (all or by port) with a user selected DSCP level?
>> > There are many guides on using regedit and other error-prone (and 
>> > geek-only) means of doing this, but is there a simple Windows 10 home app?
>> 
>> When I last tried it (years ago), in order to set the tos bits, an
>> application merely had to have admin privs.
>> 
>> > Now that Cake is out there with simple DiffServ3 support, it would be nice 
>> > to lower the priority of cloud-storage services and other bulk traffic by 
>> > correctly marking it at the origin.
>> >
>> > Cheers,
>> >
>> > Jonathan Foulkes
>> >
>> >
>> > > On Mar 15, 2019, at 3:32 PM, Jonathan Morton  
>> > > wrote:
>> > >
>> > >> On 15 Mar, 2019, at 8:36 pm, Mikael Abrahamsson  
>> > >> wrote:
>> > >>
>> > >> Having a "lower-than-best-effort" diffserve codepoint might work, 
>> > >> because it means worse treatment, not preferential treatment.
>> > >>
>> > >> The problem with having DSCP CPs that indicate preferential treatment 
>> > >> is typically a ddos magnet.
>> > >
>> > > This is true, and also why I feel that just 2 bits should be sufficient 
>> > > for Diffserv (rather than 6). They are sufficient to express four 
>> > > different optimisation targets:
>> > >
>> > > 0: Maximum Throughput (aka Best Effort)
>> > > 1: Minimum Cost (aka Least Effort)
>> > > 2: Minimum Latency (aka Maximum Responsiveness)
>> > > 3: Minimum Loss (aka Maximum Reliability)
>> > >
>> > > It is legitimate for traffic to request any of these four optimisations, 
>> > > with the explicit tradeoff of *not* necessarily getting optimisation in 
>> > > the other three dimensions.
>> > >
>> > > The old TOS spec erred in specifying 4 non-exclusive bits to express 
>> > > this, in addition to 3 bits for a telegram-office style "priority level" 
>> > > (which was very much ripe for abuse if not strictly 
>> > > admission-controlled). TOS was rightly considered a mess, but was 
>> > > replaced with Diffserv which was far too loose a spec to be useful in 
>> > > practice.
>> > >
>> > > But that's a separate topic from ECN per se.
>> > >
>> > > - Jonathan Morton
>> > >
>> > > ___
>> > > Bloat mailing list
>> > > Bloat@lists.bufferbloat.net
>> > > https://lists.bufferbloat.net/listinfo/bloat
>> >
>> > ___
>> > Bloat mailing list
>> > Bloat@lists.bufferbloat.net
>> > https://lists.bufferbloat.net/listinfo/bloat
>> 
>> 
>> 
>> -- 
>> 
>> Dave Täht
>> CTO, TekLibre, LLC
>> http://www.teklibre.com
>> Tel: 1-831-205-9740
>> 

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Michael Welzl]

Good question!  …. on Windows in particular, I’d really like to know this too.

The WebRTC Javascript API allows one to influence the DSCP, i.e. browsers 
normally can do that. Whether that’s true for all OSes, I don’t know.

Cheers,
Michael



> On Mar 16, 2019, at 12:45 AM, David P. Reed  wrote:
> 
> How many applications used by normal users have "admin" privileges? The 
> Browser? Email? FTP?
>  
>  
> -Original Message-
> From: "Dave Taht" 
> Sent: Friday, March 15, 2019 4:31pm
> To: "Jonathan Foulkes" 
> Cc: ecn-s...@lists.bufferbloat.net, "bloat" 
> Subject: Re: [Ecn-sane] [Bloat] [iccrg] Fwd: [tcpPrague] Implementation and 
> experimentation of TCP Prague/L4S hackaton at IETF104
> 
> On Fri, Mar 15, 2019 at 1:28 PM Jonathan Foulkes  
> wrote:
> >
> > All this discussion of DSCP marking brings to mind what happened on the 
> > Windows platform, where the OS had to suppress ALL DSCP marks, as app 
> > authors were trying to game the system.
> > And even if not trying to ‘game’ it, they have non-obvious reasons why they 
> > don’t mark traffic how one would expect. Example:
> >
> > I know an engineer who works at a cloud-storage solution company, and I 
> > asked why a long-standing customer request for DSCP marking (as bulk) was 
> > not implemented. His answer was they’d never do that, as that would impact 
> > benchmarks against their competitors for which service syncs faster. 
> >
> > Which brings me to a question: Is anyone aware of an easy to use Windows 
> > app that will allow the user to select an application and tell the OS to 
> > mark the traffic (all or by port) with a user selected DSCP level?
> > There are many guides on using regedit and other error-prone (and 
> > geek-only) means of doing this, but is there a simple Windows 10 home app?
> 
> When I last tried it (years ago), in order to set the tos bits, an
> application merely had to have admin privs.
> 
> > Now that Cake is out there with simple DiffServ3 support, it would be nice 
> > to lower the priority of cloud-storage services and other bulk traffic by 
> > correctly marking it at the origin.
> >
> > Cheers,
> >
> > Jonathan Foulkes
> >
> >
> > > On Mar 15, 2019, at 3:32 PM, Jonathan Morton  
> > > wrote:
> > >
> > >> On 15 Mar, 2019, at 8:36 pm, Mikael Abrahamsson  wrote:
> > >>
> > >> Having a "lower-than-best-effort" diffserve codepoint might work, 
> > >> because it means worse treatment, not preferential treatment.
> > >>
> > >> The problem with having DSCP CPs that indicate preferential treatment is 
> > >> typically a ddos magnet.
> > >
> > > This is true, and also why I feel that just 2 bits should be sufficient 
> > > for Diffserv (rather than 6). They are sufficient to express four 
> > > different optimisation targets:
> > >
> > > 0: Maximum Throughput (aka Best Effort)
> > > 1: Minimum Cost (aka Least Effort)
> > > 2: Minimum Latency (aka Maximum Responsiveness)
> > > 3: Minimum Loss (aka Maximum Reliability)
> > >
> > > It is legitimate for traffic to request any of these four optimisations, 
> > > with the explicit tradeoff of *not* necessarily getting optimisation in 
> > > the other three dimensions.
> > >
> > > The old TOS spec erred in specifying 4 non-exclusive bits to express 
> > > this, in addition to 3 bits for a telegram-office style "priority level" 
> > > (which was very much ripe for abuse if not strictly 
> > > admission-controlled). TOS was rightly considered a mess, but was 
> > > replaced with Diffserv which was far too loose a spec to be useful in 
> > > practice.
> > >
> > > But that's a separate topic from ECN per se.
> > >
> > > - Jonathan Morton
> > >
> > > ___
> > > Bloat mailing list
> > > Bloat@lists.bufferbloat.net
> > > https://lists.bufferbloat.net/listinfo/bloat
> >
> > ___
> > Bloat mailing list
> > Bloat@lists.bufferbloat.net
> > https://lists.bufferbloat.net/listinfo/bloat
> 
> 
> 
> -- 
> 
> Dave Täht
> CTO, TekLibre, LLC
> http://www.teklibre.com
> Tel: 1-831-205-9740
> ___
> Ecn-sane mailing list
> ecn-s...@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/ecn-sane
> ___
> Bloat mailing list
> Bloat@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/bloat

___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Sebastian Moeller]

Hi Dave,

On March 16, 2019 12:43:58 AM GMT+01:00, "David P. Reed"  
wrote:
>
>My point is that the argument for doing such balancing is that somehow
>ISPs at the entry points (representing somehow the goals of source and
>destination of each flow) will classify their flows correctly based on
>some criterion, and not select the option that allows them to "beat"
>all the others, which then causes them be "game theoreitically"
>incented to screw up the labeling.

One more argument for splitting the 6 dscp bits, one half for the ISP one half 
for end to end signalling...
I do neither expect ISPs to honor the intent bits (with the possible exception 
of the LE patter, there e2e and ISP goals seem aligned) and I also do not see 
any approach gain traction that completely takes the dscp bits away from the 
intermediate transports, whether one likes it or not, these effectively are 
under transport control and are actually used, so hoping the current owner 
giving all of them back seems overly optimistic. Hence the 50/50 split.


> 
>The business argument that the users at both ends will choose the rignt
>labels is that they are responsive to price signals in a very sensitive
>way that will get them to mark things correctly.

Right or wrong, or how to interpret the different pattern is a question that 
only becomes relevant once the patterns are signalled robustly end to end...

 (that includes, by the
>way, the Internet Access Providers, if they take over the labeling job
>and force their choice on their users, because they become the
>"endpoint")

Not with the split proposal, and yes end users depend on their ISP doing 
reasonable traffic management, but that seems orthogonal to dscp bit patterns 
to me.

> 
>So if pricing mechanisms don't work to incent labeling correctly, it
>does not matter that there exists an optimum that can be achieved by an
>Oracle who fully decides the settings on all packets of all protocols
>ever to be invented.
> 
>And that's why I brought up the issue of pricing and economics, which
>sadly really affect any kind of queue management.

Sure in the context of hoping the ISPs and the wider internet respecting 
endpoint set dscps that seems all applicable.

> 
>That's why pricing becomes a practical issue, and issues of "utility"
>to the users become important.
> 
>Now the other thing that is crucial is that the optimal state almost
>all of the time of every link in the network is that a utilization far
>from max capacity. The reason for this is the fact that the Internet
>(like almost all networks) is bursty and fractal. The law of large
>numbers doesn't smooth traffic volume over any timescale (that's the
>sense of fractalness here). There is no statistical smoothing of load -
>there are rare high peaks on some links but most links are
>underutilized, *if you want all the protocols currently used in the
>Internet to make users happy with minimal time-to-task-completion*
>(response time at the scale that matters for the particular user's
>needs at that moment).
> 
>So if most links are uncongested most of the time (and they should be
>if the folks maintaining the subnets are all doing their job by growing
>links that are congested to handle more traffic), then congestion
>management is only a peak load problem, and only affects things a small
>percentage of the time.

I concur with Jonathan, access links often run much closer to their limit than 
core networks, and the whole bufferbloat project demonstrated that a capable 
AQM system with mild tiering can make a saturated link still acceptable to use 
even for low latency applications...
But for ingress shaping it would be really great to have some trustworthy way 
of deducing the sender's intent, and dscps seem like a natural fit.

Best Regards
  Sebastian


> 
>This is very, very different from the typical "benchmark" case, which
>focuses only on dealing with peak loads, which are transient in the
>real world. Most "benchmarks" make the strange and unrealistic
>assumption that overload is steady state, and that users themselves
>don't give up and stop using an overloaded system.



> 
> 
>-Original Message-----
>From: "Jonathan Morton" 
>Sent: Friday, March 15, 2019 4:13pm
>To: "David P. Reed" 
>Cc: "Mikael Abrahamsson" ,
>ecn-s...@lists.bufferbloat.net, "bloat" 
>Subject: Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation
>and experimentation of TCP Prague/L4S hackaton at IETF104
>
>
>
>> On 15 Mar, 2019, at 9:44 pm, David P. Reed 
>wrote:
>> 
>> pricing (even dynamic pricing) of different qualities of service is
>unstable
>
>An interesting resul

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Dave Taht]

On Fri, Mar 15, 2019 at 9:04 PM Jonathan Morton  wrote:
>
> > On 15 Mar, 2019, at 7:01 pm, David P. Reed  wrote:
> >
> > > Next up for sce was going to be to find if dctcp could be modified to
> > > use it. Also, bittorrent.
> >
> > YES! I endorse this message.
>
> Actually, today I was still figuring out how to fit it to CUBIC.  I *think* 
> I've succeeded…
>
> https://github.com/dtaht/bufferbloat-rfcs/blob/master/sce/ELR%20Proposal%203%20(TCP).txt
>
> I recall DCTCP as being singularly built around a distinct setpoint from what 
> either Classic ECN or SCE expects.  So I think I might look at LEDBAT first.

LEDBAT as defined by the ietf is a waste of time, as is it did not deploy.

In terms of dealing with bittorrent, I recommend you look over the UDP
implementation in libutp, and in particular the embedded
implementation in "transmission", which is easier to play with first.
Along the way, getting setsockopt to set the diffserv bits right
finally for libutp would be nice.


>  - Jonathan Morton
>
> ___
> Ecn-sane mailing list
> ecn-s...@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/ecn-sane



-- 

Dave Täht
CTO, TekLibre, LLC
http://www.teklibre.com
Tel: 1-831-205-9740
___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Jonathan Morton]

> On 15 Mar, 2019, at 7:01 pm, David P. Reed  wrote:
> 
> > Next up for sce was going to be to find if dctcp could be modified to
> > use it. Also, bittorrent.
> 
> YES! I endorse this message.

Actually, today I was still figuring out how to fit it to CUBIC.  I *think* 
I've succeeded…

https://github.com/dtaht/bufferbloat-rfcs/blob/master/sce/ELR%20Proposal%203%20(TCP).txt

I recall DCTCP as being singularly built around a distinct setpoint from what 
either Classic ECN or SCE expects.  So I think I might look at LEDBAT first.

 - Jonathan Morton

___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Jonathan Morton]

> On 16 Mar, 2019, at 1:43 am, David P. Reed  wrote:
> 
> Now the other thing that is crucial is that the optimal state almost all of 
> the time of every link in the network is that a utilization far from max 
> capacity. The reason for this is the fact that the Internet (like almost all 
> networks) is bursty and fractal.

That can be said about some types of links but not others.

Last-mile links in particular are often saturated by their users' individual 
traffic for minutes or even hours at a time, especially the slower link 
technologies such as ADSL and 4G.  The user wants their hundred-gigabyte game 
update installed as soon as possible, even if they only have 10Mbps to suck it 
through, and they still want to use their connection for other things while 
they wait.  And this is not unreasonable; I do it regularly.

At peak times, ISP backhaul capacity can often be enough of a bottleneck for 
users to notice the congestion and induced latency; it is far from the case 
that all ISPs worldwide massively over-provision their networks to avoid 
routine congestion, even in modern technologically advanced nations.  There are 
remote islands where hundreds or thousands of users must share a single 
satellite or microwave uplink.  Cell towers are also a shared medium with 
decidedly finite capacity.

Only core networks, and the backhaul networks of certain particularly 
conscientious ISPs, can typically be described as congestion-free.  And that is 
why we discuss AQM and ECN in such detail in the first place; without 
congestion, they wouldn't be required.

The extent to which traffic classification is needed on particular types of 
link can be debated.  It could fairly be argued that we've done remarkably well 
without the benefit of a functioning Diffserv ecosystem, so there is no 
particular urgency to create one.  At the same time, it's worth discussing 
*why* Diffserv fails to do its intended job, and how a better system *could* be 
designed, because that may reveal lessons for the future.

I will say this: there are times, even with the benefit of everything Cake does 
for me, when I would prefer that BitTorrent traffic would automatically defer 
to other stuff (as it was supposedly designed to).  Several game updaters, 
including Wargaming.net's, use BitTorrent under the skin - opening and using 
several hundred flows in parallel, and thereby swamping any other traffic going 
to the same host.  It would be very straightforward for them to mark all that 
traffic as Minimum Cost, while their games themselves use Minimum Latency for 
battle traffic.

Minimum Cost is a natural choice for any transport using LEDBAT, or with 
similarly altruistic philosophy.

Minimum Latency is a natural choice for any application requiring realtime 
response - games, VoIP, remote shells.

Minimum Loss is a natural choice for applications involved in network control, 
where dropped packets could have a much greater than normal impact on overall 
network performance.

Maximum Throughput is a natural choice for general-purpose applications not 
fitting any of the above.

Pricing is not required.  Making the wrong choice will simply make your 
application perform badly on a loaded network, as the bottleneck link optimises 
for the thing you told it to optimise for, rather than for what you actually 
want.  That's all that's really needed.

 - Jonathan Morton

___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[David P. Reed]

How many applications used by normal users have "admin" privileges? The 
Browser? Email? FTP?
 
 
-Original Message-
From: "Dave Taht" 
Sent: Friday, March 15, 2019 4:31pm
To: "Jonathan Foulkes" 
Cc: ecn-s...@lists.bufferbloat.net, "bloat" 
Subject: Re: [Ecn-sane] [Bloat] [iccrg] Fwd: [tcpPrague] Implementation and 
experimentation of TCP Prague/L4S hackaton at IETF104



On Fri, Mar 15, 2019 at 1:28 PM Jonathan Foulkes  
wrote:
>
> All this discussion of DSCP marking brings to mind what happened on the 
> Windows platform, where the OS had to suppress ALL DSCP marks, as app authors 
> were trying to game the system.
> And even if not trying to ‘game’ it, they have non-obvious reasons why they 
> don’t mark traffic how one would expect. Example:
>
> I know an engineer who works at a cloud-storage solution company, and I asked 
> why a long-standing customer request for DSCP marking (as bulk) was not 
> implemented. His answer was they’d never do that, as that would impact 
> benchmarks against their competitors for which service syncs faster. 
>
> Which brings me to a question: Is anyone aware of an easy to use Windows app 
> that will allow the user to select an application and tell the OS to mark the 
> traffic (all or by port) with a user selected DSCP level?
> There are many guides on using regedit and other error-prone (and geek-only) 
> means of doing this, but is there a simple Windows 10 home app?

When I last tried it (years ago), in order to set the tos bits, an
application merely had to have admin privs.

> Now that Cake is out there with simple DiffServ3 support, it would be nice to 
> lower the priority of cloud-storage services and other bulk traffic by 
> correctly marking it at the origin.
>
> Cheers,
>
> Jonathan Foulkes
>
>
> > On Mar 15, 2019, at 3:32 PM, Jonathan Morton  wrote:
> >
> >> On 15 Mar, 2019, at 8:36 pm, Mikael Abrahamsson  wrote:
> >>
> >> Having a "lower-than-best-effort" diffserve codepoint might work, because 
> >> it means worse treatment, not preferential treatment.
> >>
> >> The problem with having DSCP CPs that indicate preferential treatment is 
> >> typically a ddos magnet.
> >
> > This is true, and also why I feel that just 2 bits should be sufficient for 
> > Diffserv (rather than 6). They are sufficient to express four different 
> > optimisation targets:
> >
> > 0: Maximum Throughput (aka Best Effort)
> > 1: Minimum Cost (aka Least Effort)
> > 2: Minimum Latency (aka Maximum Responsiveness)
> > 3: Minimum Loss (aka Maximum Reliability)
> >
> > It is legitimate for traffic to request any of these four optimisations, 
> > with the explicit tradeoff of *not* necessarily getting optimisation in the 
> > other three dimensions.
> >
> > The old TOS spec erred in specifying 4 non-exclusive bits to express this, 
> > in addition to 3 bits for a telegram-office style "priority level" (which 
> > was very much ripe for abuse if not strictly admission-controlled). TOS was 
> > rightly considered a mess, but was replaced with Diffserv which was far too 
> > loose a spec to be useful in practice.
> >
> > But that's a separate topic from ECN per se.
> >
> > - Jonathan Morton
> >
> > ___
> > Bloat mailing list
> > Bloat@lists.bufferbloat.net
> > https://lists.bufferbloat.net/listinfo/bloat
>
> ___
> Bloat mailing list
> Bloat@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/bloat



-- 

Dave Täht
CTO, TekLibre, LLC
http://www.teklibre.com
Tel: 1-831-205-9740
___
Ecn-sane mailing list
ecn-s...@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/ecn-sane___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[David P. Reed]

My point is that the argument for doing such balancing is that somehow ISPs at 
the entry points (representing somehow the goals of source and destination of 
each flow) will classify their flows correctly based on some criterion, and not 
select the option that allows them to "beat" all the others, which then causes 
them be "game theoreitically" incented to screw up the labeling.
 
The business argument that the users at both ends will choose the rignt labels 
is that they are responsive to price signals in a very sensitive way that will 
get them to mark things correctly. (that includes, by the way, the Internet 
Access Providers, if they take over the labeling job and force their choice on 
their users, because they become the "endpoint")
 
So if pricing mechanisms don't work to incent labeling correctly, it does not 
matter that there exists an optimum that can be achieved by an Oracle who fully 
decides the settings on all packets of all protocols ever to be invented.
 
And that's why I brought up the issue of pricing and economics, which sadly 
really affect any kind of queue management.
 
That's why pricing becomes a practical issue, and issues of "utility" to the 
users become important.
 
Now the other thing that is crucial is that the optimal state almost all of the 
time of every link in the network is that a utilization far from max capacity. 
The reason for this is the fact that the Internet (like almost all networks) is 
bursty and fractal. The law of large numbers doesn't smooth traffic volume over 
any timescale (that's the sense of fractalness here). There is no statistical 
smoothing of load - there are rare high peaks on some links but most links are 
underutilized, *if you want all the protocols currently used in the Internet to 
make users happy with minimal time-to-task-completion* (response time at the 
scale that matters for the particular user's needs at that moment).
 
So if most links are uncongested most of the time (and they should be if the 
folks maintaining the subnets are all doing their job by growing links that are 
congested to handle more traffic), then congestion management is only a peak 
load problem, and only affects things a small percentage of the time.
 
This is very, very different from the typical "benchmark" case, which focuses 
only on dealing with peak loads, which are transient in the real world. Most 
"benchmarks" make the strange and unrealistic assumption that overload is 
steady state, and that users themselves don't give up and stop using an 
overloaded system.
 
 
-Original Message-
From: "Jonathan Morton" 
Sent: Friday, March 15, 2019 4:13pm
To: "David P. Reed" 
Cc: "Mikael Abrahamsson" , ecn-s...@lists.bufferbloat.net, 
"bloat" 
Subject: Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and 
experimentation of TCP Prague/L4S hackaton at IETF104



> On 15 Mar, 2019, at 9:44 pm, David P. Reed  wrote:
> 
> pricing (even dynamic pricing) of different qualities of service is unstable

An interesting result, but I should note that the four-way optimisation system 
I described doesn't rely on pricing, only a sufficiently correct implementation 
of those optimisations at enough bottlenecks to make it worthwhile for 
applications to mark their traffic appropriately. The technology exists to do 
so, but is not standardised in a way that makes it usable.

 - Jonathan Morton

___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Dave Taht]

On Fri, Mar 15, 2019 at 1:28 PM Jonathan Foulkes  
wrote:
>
> All this discussion of DSCP marking brings to mind what happened on the 
> Windows platform, where the OS had to suppress ALL DSCP marks, as app authors 
> were trying to game the system.
> And even if not trying to ‘game’ it, they have non-obvious reasons why they 
> don’t mark traffic how one would expect. Example:
>
> I know an engineer who works at a cloud-storage solution company, and I asked 
> why a long-standing customer request for DSCP marking (as bulk) was not 
> implemented. His answer was they’d never do that, as that would impact 
> benchmarks against their competitors for which service syncs faster. 
>
> Which brings me to a question: Is anyone aware of an easy to use Windows app 
> that will allow the user to select an application and tell the OS to mark the 
> traffic (all or by port) with a user selected DSCP level?
> There are many guides on using regedit and other error-prone (and geek-only) 
> means of doing this, but is there a simple Windows 10 home app?

When I last tried it (years ago), in order to set the tos bits, an
application merely had to have admin privs.

> Now that Cake is out there with simple DiffServ3 support, it would be nice to 
> lower the priority of cloud-storage services and other bulk traffic by 
> correctly marking it at the origin.
>
> Cheers,
>
> Jonathan Foulkes
>
>
> > On Mar 15, 2019, at 3:32 PM, Jonathan Morton  wrote:
> >
> >> On 15 Mar, 2019, at 8:36 pm, Mikael Abrahamsson  wrote:
> >>
> >> Having a "lower-than-best-effort" diffserve codepoint might work, because 
> >> it means worse treatment, not preferential treatment.
> >>
> >> The problem with having DSCP CPs that indicate preferential treatment is 
> >> typically a ddos magnet.
> >
> > This is true, and also why I feel that just 2 bits should be sufficient for 
> > Diffserv (rather than 6).  They are sufficient to express four different 
> > optimisation targets:
> >
> > 0: Maximum Throughput (aka Best Effort)
> > 1: Minimum Cost (aka Least Effort)
> > 2: Minimum Latency (aka Maximum Responsiveness)
> > 3: Minimum Loss (aka Maximum Reliability)
> >
> > It is legitimate for traffic to request any of these four optimisations, 
> > with the explicit tradeoff of *not* necessarily getting optimisation in the 
> > other three dimensions.
> >
> > The old TOS spec erred in specifying 4 non-exclusive bits to express this, 
> > in addition to 3 bits for a telegram-office style "priority level" (which 
> > was very much ripe for abuse if not strictly admission-controlled).  TOS 
> > was rightly considered a mess, but was replaced with Diffserv which was far 
> > too loose a spec to be useful in practice.
> >
> > But that's a separate topic from ECN per se.
> >
> > - Jonathan Morton
> >
> > ___
> > Bloat mailing list
> > Bloat@lists.bufferbloat.net
> > https://lists.bufferbloat.net/listinfo/bloat
>
> ___
> Bloat mailing list
> Bloat@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/bloat



-- 

Dave Täht
CTO, TekLibre, LLC
http://www.teklibre.com
Tel: 1-831-205-9740
___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Jonathan Foulkes]

All this discussion of DSCP marking brings to mind what happened on the Windows 
platform, where the OS had to suppress ALL DSCP marks, as app authors were 
trying to game the system.
And even if not trying to ‘game’ it, they have non-obvious reasons why they 
don’t mark traffic how one would expect. Example:

I know an engineer who works at a cloud-storage solution company, and I asked 
why a long-standing customer request for DSCP marking (as bulk) was not 
implemented. His answer was they’d never do that, as that would impact 
benchmarks against their competitors for which service syncs faster. 

Which brings me to a question: Is anyone aware of an easy to use Windows app 
that will allow the user to select an application and tell the OS to mark the 
traffic (all or by port) with a user selected DSCP level?
There are many guides on using regedit and other error-prone (and geek-only) 
means of doing this, but is there a simple Windows 10 home app?

Now that Cake is out there with simple DiffServ3 support, it would be nice to 
lower the priority of cloud-storage services and other bulk traffic by 
correctly marking it at the origin. 

Cheers,

Jonathan Foulkes


> On Mar 15, 2019, at 3:32 PM, Jonathan Morton  wrote:
> 
>> On 15 Mar, 2019, at 8:36 pm, Mikael Abrahamsson  wrote:
>> 
>> Having a "lower-than-best-effort" diffserve codepoint might work, because it 
>> means worse treatment, not preferential treatment.
>> 
>> The problem with having DSCP CPs that indicate preferential treatment is 
>> typically a ddos magnet.
> 
> This is true, and also why I feel that just 2 bits should be sufficient for 
> Diffserv (rather than 6).  They are sufficient to express four different 
> optimisation targets:
> 
> 0: Maximum Throughput (aka Best Effort)
> 1: Minimum Cost (aka Least Effort)
> 2: Minimum Latency (aka Maximum Responsiveness)
> 3: Minimum Loss (aka Maximum Reliability)
> 
> It is legitimate for traffic to request any of these four optimisations, with 
> the explicit tradeoff of *not* necessarily getting optimisation in the other 
> three dimensions.
> 
> The old TOS spec erred in specifying 4 non-exclusive bits to express this, in 
> addition to 3 bits for a telegram-office style "priority level" (which was 
> very much ripe for abuse if not strictly admission-controlled).  TOS was 
> rightly considered a mess, but was replaced with Diffserv which was far too 
> loose a spec to be useful in practice.
> 
> But that's a separate topic from ECN per se.
> 
> - Jonathan Morton
> 
> ___
> Bloat mailing list
> Bloat@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/bloat

___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Jonathan Morton]

> On 15 Mar, 2019, at 9:44 pm, David P. Reed  wrote:
> 
> pricing (even dynamic pricing) of different qualities of service is unstable

An interesting result, but I should note that the four-way optimisation system 
I described doesn't rely on pricing, only a sufficiently correct implementation 
of those optimisations at enough bottlenecks to make it worthwhile for 
applications to mark their traffic appropriately.  The technology exists to do 
so, but is not standardised in a way that makes it usable.

 - Jonathan Morton

___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[David P. Reed]

Just to throw in one more thing not well understood by engineers.
 
Economists I have discussed this with (real ones, not fringe right-wing true 
believers that the market "just works"), have observed that pricing (even 
dynamic pricing) of different qualities of service is unstable and extremely 
unlikely to reflect the correct price for the particular utility of the 
achieved service quality.
 
The point of that observation is that even a simple 2 classes of service system 
(so-called Paris Metro Pricing) is unstable, such that users of such a system 
will not be encouraged to set the priorities/service types to make system 
optimal or stable.
 
I can explain more, but the end user doesn't benefit from multiple choices of 
class of service at the packet level.
-Original Message-
From: "Jonathan Morton" 
Sent: Friday, March 15, 2019 3:32pm
To: "Mikael Abrahamsson" 
Cc: "David P. Reed" , ecn-s...@lists.bufferbloat.net, 
"bloat" 
Subject: Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and 
experimentation of TCP Prague/L4S hackaton at IETF104



> On 15 Mar, 2019, at 8:36 pm, Mikael Abrahamsson  wrote:
> 
> Having a "lower-than-best-effort" diffserve codepoint might work, because it 
> means worse treatment, not preferential treatment.
> 
> The problem with having DSCP CPs that indicate preferential treatment is 
> typically a ddos magnet.

This is true, and also why I feel that just 2 bits should be sufficient for 
Diffserv (rather than 6). They are sufficient to express four different 
optimisation targets:

0: Maximum Throughput (aka Best Effort)
1: Minimum Cost (aka Least Effort)
2: Minimum Latency (aka Maximum Responsiveness)
3: Minimum Loss (aka Maximum Reliability)

It is legitimate for traffic to request any of these four optimisations, with 
the explicit tradeoff of *not* necessarily getting optimisation in the other 
three dimensions.

The old TOS spec erred in specifying 4 non-exclusive bits to express this, in 
addition to 3 bits for a telegram-office style "priority level" (which was very 
much ripe for abuse if not strictly admission-controlled). TOS was rightly 
considered a mess, but was replaced with Diffserv which was far too loose a 
spec to be useful in practice.

But that's a separate topic from ECN per se.

 - Jonathan Morton

___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Jonathan Morton]

> On 15 Mar, 2019, at 8:36 pm, Mikael Abrahamsson  wrote:
> 
> Having a "lower-than-best-effort" diffserve codepoint might work, because it 
> means worse treatment, not preferential treatment.
> 
> The problem with having DSCP CPs that indicate preferential treatment is 
> typically a ddos magnet.

This is true, and also why I feel that just 2 bits should be sufficient for 
Diffserv (rather than 6).  They are sufficient to express four different 
optimisation targets:

0: Maximum Throughput (aka Best Effort)
1: Minimum Cost (aka Least Effort)
2: Minimum Latency (aka Maximum Responsiveness)
3: Minimum Loss (aka Maximum Reliability)

It is legitimate for traffic to request any of these four optimisations, with 
the explicit tradeoff of *not* necessarily getting optimisation in the other 
three dimensions.

The old TOS spec erred in specifying 4 non-exclusive bits to express this, in 
addition to 3 bits for a telegram-office style "priority level" (which was very 
much ripe for abuse if not strictly admission-controlled).  TOS was rightly 
considered a mess, but was replaced with Diffserv which was far too loose a 
spec to be useful in practice.

But that's a separate topic from ECN per se.

 - Jonathan Morton

___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Sebastian Moeller]

Hi Mikael,


> On Mar 15, 2019, at 19:36, Mikael Abrahamsson  wrote:
> 
> On Fri, 15 Mar 2019, David P. Reed wrote:
> 
>> So if the responsible network engineers in the carriers cannot agree on 
>> anything, IETF is wasting its time.
> 
> The IETF has already said that anything diffserv is domain-internal only. I 
> have joined the effort of the LE PHB and see if we can get some kind of 
> agreement and transparancy for a PHB that is aimed at customer access only 
> and "drop most of me and my pals at any sign of customer access line 
> congestion", and see if that can be agreed on.

+1

> 
> Having a "lower-than-best-effort" diffserve codepoint might work, because it 
> means worse treatment, not preferential treatment.
> 
> The problem with having DSCP CPs that indicate preferential treatment is 
> typically a ddos magnet.

Hence splitting it up, three for the current transport domain to do 
with as it sees fit and 3 for signaling intent; this very much does not give a 
guarantee that any intermediate hop will follow the intent, but only make it 
possible for the endpoints to transmit intent. This IMHO is completely 
compatible with a LE PHB and transports honoring that request. 

> See my emails on this topic on (this? other?) mailing lists where I try to 
> create a three class buffering system saying "LE gets 5%, BE and 
> 'everything-else' gets to split the difference".

We can haggle over the numbers but that seems a) sane and b) 
underspecified...

> 
> I even got pushback on this here, and then we're not even close to people 
> running large ISP networks who see ddos attacks happen hourly.
> 
> Saying L4S should "just use diffserv" is as constructive to say "go away and 
> pound a rock" or "we want that bit pattern so.. screw you".

But just nodding expertly when they go and claim an unrelated bit in 
the IP header for their separation l4s vs legacy (as if l4s would be the end 
all of network design), and then having resorting to modifying so-far 
not-deployed-at the edge DCTCP (instead of modifying well-deployed TCP) because 
they already spent the one bit usable to extend ECN for less binary congestion 
signaling in a backward-compatible fashion... I might be wording things to 
strongly here, but that is the general gist.

> 
> L4S has a much better possibility of actually getting deployment into the 
> wider Internet packet-moving equipment than anything being talked about here.

That is not a high bar to clear though...

> Same with PIE as opposed to FQ_CODEL. I know it's might not be as good,

Debatable, and from my perspective this is the reason to talk about it 
at all.

> but it fits better into actual silicon

Does it?

> and it's being proposed by people who actually have better channels into the 
> people setting hard requirements.

That would be great if the proposal would throw end-user like me a bone 
instead of treating me as the product. It would also help if the architectural 
RFC would not be so breathlessly over-hyping/over-promising... But they really 
need end-points to switch over to a neutered DCTCP before things start to make 
sense, so they actually need to convince end-users and so far they are doing a 
terrible job IMHO. But what do I know...

> 
> I suggest you consider joining them instead of opposing them.

Join where? it pretty much looks like a "fait accompli" as they do seem 
way past the design stages and seem pretty much crystallized in what they see 
the path forward. 

Best Regards
Sebastian

> 
> -- 
> Mikael Abrahamssonemail: swm...@swm.pp.se

___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Mikael Abrahamsson]

On Fri, 15 Mar 2019, David P. Reed wrote:

So if the responsible network engineers in the carriers cannot agree on 
anything, IETF is wasting its time.


The IETF has already said that anything diffserv is domain-internal only. 
I have joined the effort of the LE PHB and see if we can get some kind of 
agreement and transparancy for a PHB that is aimed at customer access only 
and "drop most of me and my pals at any sign of customer access line 
congestion", and see if that can be agreed on.


Having a "lower-than-best-effort" diffserve codepoint might work, because 
it means worse treatment, not preferential treatment.


The problem with having DSCP CPs that indicate preferential treatment is 
typically a ddos magnet. See my emails on this topic on (this? other?) 
mailing lists where I try to create a three class buffering system saying 
"LE gets 5%, BE and 'everything-else' gets to split the difference".


I even got pushback on this here, and then we're not even close to people 
running large ISP networks who see ddos attacks happen hourly.


Saying L4S should "just use diffserv" is as constructive to say "go away 
and pound a rock" or "we want that bit pattern so.. screw you".


L4S has a much better possibility of actually getting deployment into the 
wider Internet packet-moving equipment than anything being talked about 
here. Same with PIE as opposed to FQ_CODEL. I know it's might not be as 
good, but it fits better into actual silicon and it's being proposed by 
people who actually have better channels into the people setting hard 
requirements.


I suggest you consider joining them instead of opposing them.

--
Mikael Abrahamssonemail: swm...@swm.pp.se
___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Mikael Abrahamsson]

On Fri, 15 Mar 2019, Dave Taht wrote:

It appears, also, ironically, (I have confirmation from several sources 
now) that cake, fq_codel and dualpi are now illegal for an ISP to use in 
their provided equipment under california law. The idea of one day 
having to appear in court to defend our key algorithms reminds me of the 
famous john fogerty case where he demonstrated how blues music was made.


I would love to know more about this. Running an AQM on the customer 
access that doesn't prioritize some specific service should be fine, ie it 
doesn't explicitly do something the *customer* doesn't benefit from.


Net neutrality should be about what the ISP does to benefit itself, not 
what is done on the customer access line (the scheduler per customer) that 
just looks at packet flow characteristics and isn't configured to 
prioritize any specific content (src/destination/port etc).


--
Mikael Abrahamssonemail: swm...@swm.pp.se
___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[Sebastian Moeller]

On March 15, 2019 6:01:23 PM GMT+01:00, "David P. Reed"  
wrote:
>
>The absolute fundamental issue with diffserv, IMO, is that the carriers
>cannot agree on an actual specification of what routers and gateways
>are supposed to do, while being compatible with the core principle of
>the IP layer: do not hold packets if they cause increasing queueing
>delay. (this is in the Best Practices RFC)
> 

IMHO it is the Charme of the 2 times 3 bits approach, that carriers get 3 bits 
they can do with whatever they want. As VLAN tags and MPLS? only support 3 bit 
priorities this looks to me like a match made in heaven, and we get 3 bits with 
end to end guarantees. Not that rolling that out would ever work in reality

Best Regards

Sebastian

And yes, this is not an idea I came up with, I just forgot who proposed that 
initially.


>And it's not for lack of trying. Dave Clark led a working group at the
>MIT communications futures program (where I was a principle) that
>included most major backbone providers' key folks that was specifically
>focused on getting a widespread agreement on what any of the code
>points might mean, not as bullshit English descriptions of what kind of
>traffic each one represented, but as an operational description of what
>should be done by a router to manage its queues.
> 
>They couldn't even agree (after about 18 months of meetings) about what
>the bullshit English intent was, much less what operational semantics
>in the packet forwarding network had to be.
> 
>So if the responsible network engineers in the carriers cannot agree on
>anything, IETF is wasting its time.
> 
> 
> 
> 
>-Original Message-
>From: "Sebastian Moeller" 
>Sent: Friday, March 15, 2019 11:52am
>To: "Dave Täht" 
>Cc: ecn-s...@lists.bufferbloat.net, "bloat"
>
>Subject: Re: [Ecn-sane] [Bloat] [iccrg] Fwd: [tcpPrague] Implementation
>and experimentation of TCP Prague/L4S hackaton at IETF104
>
>
>
>Hi Dave,
>
>
>> On Mar 15, 2019, at 15:06, Dave Taht  wrote:
>> 
>> I would really prefer to move this discussion to the ecn-sane mailing
>> list, as IMHO, ecn is generally such a tiny thing needed for good
>> congestion control compared to better transports like pacing + bbr,
>> and things like bql, fq, and aqm using drop.
>> 
>> I'm going to keep cc-ing that list in the hope that we can keep
>better
>> track of the discussion here.
>
>Ah, sure, I basically wanted to avoid annoying the ietf lists as I feel
>out of place there, having only had a cursory read of the relevant
>RFCs.
>
>
>> 
>> On Fri, Mar 15, 2019 at 6:01 AM Sebastian Moeller 
>wrote:
>>> 
>>> Hi Dave,
>>> 
>>> I pruned the CC list as I am out of my league here and want to
>restrict the radius of my embarrassment to those that already know my
>level of incompetence before hand.
>> 
>> IMHO, your work on educating the OpenWrt community over the years on
>> how to use sqm, makes you much more than "only a grasshopper". You
>> have a firm grip on what can be achieved in the real world.
>> 
>>> 
>>> That said, having read through the L4S architecture description and
>the related appendices of draft-ietf-tsvwg-ecn-l4s-id-05 I came to the
>conclusion, that this is a mess.
>> 
>> I am so glad someone other than I has now read it.
>> 
>>> The L4S project proposes a really wide-ranging change of basically
>the internet (but allow a concurrent operation with legacy probably to
>cater to the fact that an internet-wide flag-day seems daunting to
>organize). But then they chicken out when figuring out how to
>differentiate between their new and the old by proposing to use ECT(1)
>for a purpose outside of its nominal purpose namely explicit congestion
>notification, why not think bolder? If the plan is to change everything
>the feasibility can not hinge upon the ability to re-using one old
>legacy bit, can it...
>>> Conceptually it seems much cleaner to use ECT(1) for congestion
>notification directly (there are already proposals out there and SCE
>just added another fully back-ward compatible one) and find another way
>to mark l4s traffic, sure that is going to be hard and inconvenient,
>but if you set out to change the internet that is par for the course.
>>> IMHO they would do more good if they actually fought for a better
>use of the 6 DSCP bits instead. (say by splitting into two groups of 3
>bits, one group for reduced diffserv and one group for new features,
>that would even
>> 
>> The existing diffserv deployment is a failure.
>
> Which is a shame, but one RFC's failure is another one's opportunity.
>
>
>> I have another ID
>> cooking that suggests a better way, going forward, to use them, but I
>> do not feel at this time it would be useful to present. One big
>battle
>> at a time.
>
>:)
>
>> That ID is very simple, it basically proposes that all
>> diffserv codepoints be passed through ISPs and transit providers
>> unchanged, and if any given codepoint must be changed, the only
>> permissible change is to 0 (BE), and MUST be not be remarked to
>> anything

Re: [Bloat] [Ecn-sane] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

[David P. Reed]

The absolute fundamental issue with diffserv, IMO, is that the carriers cannot 
agree on an actual specification of what routers and gateways are supposed to 
do, while being compatible with the core principle of the IP layer: do not hold 
packets if they cause increasing queueing delay. (this is in the Best Practices 
RFC)
 
And it's not for lack of trying. Dave Clark led a working group at the MIT 
communications futures program (where I was a principle) that included most 
major backbone providers' key folks that was specifically focused on getting a 
widespread agreement on what any of the code points might mean, not as bullshit 
English descriptions of what kind of traffic each one represented, but as an 
operational description of what should be done by a router to manage its queues.
 
They couldn't even agree (after about 18 months of meetings) about what the 
bullshit English intent was, much less what operational semantics in the packet 
forwarding network had to be.
 
So if the responsible network engineers in the carriers cannot agree on 
anything, IETF is wasting its time.
 
 
 
 
-Original Message-
From: "Sebastian Moeller" 
Sent: Friday, March 15, 2019 11:52am
To: "Dave Täht" 
Cc: ecn-s...@lists.bufferbloat.net, "bloat" 
Subject: Re: [Ecn-sane] [Bloat] [iccrg] Fwd: [tcpPrague] Implementation and 
experimentation of TCP Prague/L4S hackaton at IETF104



Hi Dave,


> On Mar 15, 2019, at 15:06, Dave Taht  wrote:
> 
> I would really prefer to move this discussion to the ecn-sane mailing
> list, as IMHO, ecn is generally such a tiny thing needed for good
> congestion control compared to better transports like pacing + bbr,
> and things like bql, fq, and aqm using drop.
> 
> I'm going to keep cc-ing that list in the hope that we can keep better
> track of the discussion here.

 Ah, sure, I basically wanted to avoid annoying the ietf lists as I feel out of 
place there, having only had a cursory read of the relevant RFCs.


> 
> On Fri, Mar 15, 2019 at 6:01 AM Sebastian Moeller  wrote:
>> 
>> Hi Dave,
>> 
>> I pruned the CC list as I am out of my league here and want to restrict the 
>> radius of my embarrassment to those that already know my level of 
>> incompetence before hand.
> 
> IMHO, your work on educating the OpenWrt community over the years on
> how to use sqm, makes you much more than "only a grasshopper". You
> have a firm grip on what can be achieved in the real world.
> 
>> 
>> That said, having read through the L4S architecture description and the 
>> related appendices of draft-ietf-tsvwg-ecn-l4s-id-05 I came to the 
>> conclusion, that this is a mess.
> 
> I am so glad someone other than I has now read it.
> 
>> The L4S project proposes a really wide-ranging change of basically the 
>> internet (but allow a concurrent operation with legacy probably to cater to 
>> the fact that an internet-wide flag-day seems daunting to organize). But 
>> then they chicken out when figuring out how to differentiate between their 
>> new and the old by proposing to use ECT(1) for a purpose outside of its 
>> nominal purpose namely explicit congestion notification, why not think 
>> bolder? If the plan is to change everything the feasibility can not hinge 
>> upon the ability to re-using one old legacy bit, can it...
>> Conceptually it seems much cleaner to use ECT(1) for congestion notification 
>> directly (there are already proposals out there and SCE just added another 
>> fully back-ward compatible one) and find another way to mark l4s traffic, 
>> sure that is going to be hard and inconvenient, but if you set out to change 
>> the internet that is par for the course.
>> IMHO they would do more good if they actually fought for a better use of the 
>> 6 DSCP bits instead. (say by splitting into two groups of 3 bits, one group 
>> for reduced diffserv and one group for new features, that would even
> 
> The existing diffserv deployment is a failure.

 Which is a shame, but one RFC's failure is another one's opportunity.


> I have another ID
> cooking that suggests a better way, going forward, to use them, but I
> do not feel at this time it would be useful to present. One big battle
> at a time.

:)

> That ID is very simple, it basically proposes that all
> diffserv codepoints be passed through ISPs and transit providers
> unchanged, and if any given codepoint must be changed, the only
> permissible change is to 0 (BE), and MUST be not be remarked to
> anything else, especially not CS1.

 I like the simplicity, but I also like the split into two groups of 3 bits, 
say "active bit pattern" (for transport purposes) and "intenden bit pattern" 
for the sender to transmit intent, which than can be conveyed lossless to the 
receiver; my gut feeling is that throwing the transport people a bone here 
might work better, as in the end they are the one that will carry the "burden" 
of the change. IMHO this has the additional advantage that it will make "tabula 
rasa" of the existing distict set of