Re: [tsvwg] [iccrg] SCReAM (RFC8298) with CoDel-ECN and L4S

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

Hi Bob

Please see inline 

 

/Ingemar

 

From: Bob Briscoe <[e-mail]> 
Sent: den 17 mars 2020 02:01
To: Ingemar Johansson S <ingemar.s.johansson=[e-mail]>
Cc: Sebastian Moeller <[e-mail]>; Ingemar Johansson S <[e-mail]>; [e-mail]; [e-mail]
Subject: Re: [iccrg] [tsvwg] SCReAM (RFC8298) with CoDel-ECN and L4S

 

Ingemar,

On 10/03/2020 12:37, Ingemar Johansson S wrote:

Hi
 
The SCReAM code is freely available on https://github.com/EricssonResearch/scream <https://protect2.fireeye.com/v1/url?k=70bef122-2c35fa1d-70beb1b9-868f25761b72-27b51935b33618c8&q=1&e=6939be47-9753-40b0-819b-9c42d21fc6e4&u=https%3A%2F%2Fgithub.com%2FEricssonResearch%2Fscream>  for anybody interested to run their own experiment with whatever AQM/ECN configuration. 
Please note that SCReAM is configured in an L4S mode when the network AQM does L4S marking (mimicking ECT(1)). For CoDel-ECN however, SCReAM runs in normal ECN mode with a beta of 0.8 (=20% reduction on CWND for each congestion event)


[BB] In your previous work, in order to get early L4S ECN marking, you used a virtual queue in the network draining at a little less than the rate given by the scheduler. Am I right that you're not doing that here? 

[IJ] That is correct and yes, virtual queues are definitely a possibility to achieve zero (or near zero) queue delay.



In the L4S case you show, when the video is running at about half the rate in the CoDel case (which is presumably closer to that available from the scheduler), one would have thought it would get very little ECN marking. But I notice it's getting frequent little spikes of marking. Is that due to I-frames causing bursts or something?

What if it wasn't getting those spikes of marks (e.g. using sthg like x264 to spread the I-frames over time)?

[IJ] If periodic intra refresh was used instead then the “peak to average ratio” (for lack of a better term) in frame sizes would decrease and SCReAM would increase its target bitrate. Periodic intra refresh is however not always feasible, for instance scene changes (in gaming to take one example) will require IDR frames. One question mark here is what impact an occasional large I-frame will have then?. In the simple bottleneck case here, there is no more capacity available and one will inevitably get a delay spike. In a e.g 5G system there is a possibility to use different scheduling algorithms that makes it possible for flows to temporarily scavenge on the available capacity. I don’t however have the data to show that yet.   



A random thought: Could the video use the Classic queue as a scavenger service to fill in extra coding layer(s) whenever bandwidth is available?

[IJ] Depends on the delay budget, I guess. 




Bob





 
 
I tried also with other different ramp markers (1ms/10ms), (2ms/10ms),(5ms/15ms). There are slight variations  in throughput and latency but not dramatic.
And truth to be told, the ECN behavior is better tuned in the code than the L4S behavior. 
There is room for improvement as regards to the L4S behavior (for instance faster ramp-up) and it may well be the case that SCReAM is completely scrapped in favor of new designs. 
But the bottomline, the L4S thresholds and L4S code is not carefully picked to show a good performance.
 
/Ingemar
 

-----Original Message-----
From: Sebastian Moeller  <mailto:[e-mail]> <[e-mail]>
Sent: den 10 mars 2020 11:28
To: Ingemar Johansson S  <mailto:[e-mail]> <[e-mail]>
Cc: Ingemar Johansson S
 <mailto:ingemar.s.johansson=[e-mail]> <ingemar.s.johansson=[e-mail]>; [e-mail] <mailto:[e-mail]> ;
[e-mail] <mailto:[e-mail]> 
Subject: Re: [tsvwg] SCReAM (RFC8298) with CoDel-ECN and L4S
 
Hi Ingemar,
 
 

On Mar 10, 2020, at 11:07, Ingemar Johansson S

 <mailto:[e-mail]> <[e-mail]> wrote:

 
Hi
 
For the future studies we will only focus on L4S as the scope is to study the

performance gain that L4S give for instance for AR/VR, gaming and remote
control applications.
 
   [SM] How are you going to "study the performance gain that L4S
give[s]" if you do not compare it with the best of class alternatives? I am truly
puzzled.
 

Flow aware AQMs with RTT estimates as metadata in the packets is outside

the scope as it would require packet inspection, which is not feasible if queues
build up on the RLC layer in the 3GPP stack.
 
   [SM] Fair enough. What is this comparison intended to show us then?
 
As far as I can see you paired an application designed for 1/p-type congestion
feed-back with an 1/sqrt(p)-type AQM that was also set for sub-optimal RTT and
latency target for the test path. And lo and behold, the application does
"better*" for the 1/p-type AQM (with lower latency target; I assume that  L4S
ramp-marker (Th_low=2ms, Th_high=10ms) was carefully selected to match
what SCReAM expects). IMHO that simply demonstrates, that in communication
it pays if sender and receiver of a symbol (CE here) assign the same meaning to
it.
 
But that can not be it, sohat am I missing here?
 
Best Regards
   Sebastian
 
 
*) Assuming one buys into your definition of better, in which (instantaneous)
queueing delay is valued over video quality. From a network operators
perspective that seems a valid position
 
 
 
 

 
/Ingemar
 

-----Original Message-----
From: Sebastian Moeller  <mailto:[e-mail]> <[e-mail]>
Sent: den 10 mars 2020 10:45
To: Ingemar Johansson S
 <mailto:ingemar.s.johansson=[e-mail]> <ingemar.s.johansson=[e-mail]>
Cc: [e-mail] <mailto:[e-mail]> ; Ingemar Johansson S
 <mailto:[e-mail]> <[e-mail]>; [e-mail] <mailto:[e-mail]> 
Subject: Re: [tsvwg] SCReAM (RFC8298) with CoDel-ECN and L4S
 
Hi Ingemar,
 
thanks for posting this interesting piece of data!
 

On Mar 10, 2020, at 09:02, Ingemar Johansson S

 <mailto:ingemar.s.johansson=[e-mail]> <ingemar.s.johansson=[e-mail]> wrote:

 
Hi
 
I recently updated the readme on the SCReAM github with a comparison
with

SCReAM in three different settings

       • No ECN
       • CoDel ECN
       • L4S
https://protect2.fireeye.com/v1/url?k=63019d27-3f884737-6301ddbc-0cc
47
ad93e2a-489fa99c3277fb8a&q=1&e=5aab95a7-4aab-4a64-99a5-

5b55606e303b&u=

https%3A%2F%2Fgithub.com%2FEricssonResearch%2Fscream%23ecn-

explicit-

co

ngestion-notification
 
Even though it is more than a magnitude difference in queue delay
between CoDel-ECN and L4S,

 
 
      [SM] So, in this simulations of a 20ms path, SCReAM over L4S gives
~10 times less queueing delay, but also only ~2 less bandwidth
compared to SCReAM over codel. You describe this as "L4S reduces the
delay considerably more" and "L4S gives a somewhat lower media rate".
I wonder how many end-users would tradeoff these 25ms in queueing
delay against the decrease in video quality from halving the bitrate?
Could you repeat the Codel test with interval set to 20 and target to
1ms, please?
 
If that improves things considerably it would argue for embedding the
current best RTT estimate into SCReAM packets, so an AQM could tailor
its signaling better to individual flow properties (and yes, that
will require a flow-aware AQM).
 
 
 

it is fair to say that these simple simulations should of course be
seen as just a

snapshot.
 
      [SM] Fair enough.
 

We hope to present some more simulations with 5G access, and not
just

simple bottlenecks with one flow, after the summer.
 
      [Looking] forward to that.
 

 
Meanwhile, the SCReAM code on github is freely available for anyone
who

wish to make more experiments.

 
/Ingemar
================================
Ingemar Johansson  M.Sc.
Master Researcher
 
Ericsson Research
RESEARCHER
GFTL ER NAP NCM Netw Proto & E2E Perf Labratoriegränd 11
971 28, Luleå, Sweden
Phone [phone]
SMS/MMS [phone]
[e-mail] <mailto:[e-mail]> 
www.ericsson.com <http://www.ericsson.com> 
 
 Reality, is the only thing… That’s real!
     James Halliday, Ready Player One
=================================

 





_______________________________________________
iccrg mailing list
[e-mail] <mailto:[e-mail]> 
https://www.irtf.org/mailman/listinfo/iccrg <https://protect2.fireeye.com/v1/url?k=4fa6ca14-132dc12b-4fa68a8f-868f25761b72-1432ef259e224229&q=1&e=6939be47-9753-40b0-819b-9c42d21fc6e4&u=https%3A%2F%2Fwww.irtf.org%2Fmailman%2Flistinfo%2Ficcrg> 





-- 
________________________________________________________________
Bob Briscoe                               http://bobbriscoe.net/ <https://protect2.fireeye.com/v1/url?k=4b6f9fdb-17e494e4-4b6fdf40-868f25761b72-0cf56bc001a75473&q=1&e=6939be47-9753-40b0-819b-9c42d21fc6e4&u=http%3A%2F%2Fbobbriscoe.net%2F>