Re: [Bloat] [EXTERNAL] Re: [Starlink] Researchers Seeking Probe Volunteers in USA

2023-01-09 Thread Dave Taht via Bloat
On Mon, Jan 9, 2023 at 10:54 AM Livingood, Jason
 wrote:
>
> > 0) None of the tests last long enough.
>
> The user-initiated ones tend to be shorter - likely because the average user 
> does not want to wait several minutes for a test to complete. But IMO this is 
> where a test platform like SamKnows, Ookla's embedded client, NetMicroscope, 
> and others can come in - since they run in the background on some randomized 
> schedule w/o user intervention. Thus, the user's time-sensitivity is no 
> longer a factor and a longer duration test can be performed.

I would be so happy if someone independent ( and not necessarily me!)
was validating those more private tests, and retaining reference
packet captures of the various behaviors observed.

Bloat is just one problem among many, that shows up on a speedtest. I
have, for example, been working for months, on a very difficult
problem occurring on at least one wifi6 chipset... where the block-ack
window is being violated, leading to a ton of jitter under certain
loads, and a bandwidth reduction, that doesn't show up in summary
data.

Samknows published a really good blog recently,
here: https://samknows.com/blog/testing-principles

about how they are going about things... however...

>
> > 1) Not testing up + down + ping at the same time
>
> You should consider publishing a LUL BCP I-D in the IRTF/IETF - like in 
> IPPM...

I have cc'd ippm a few times on these threads, and am on that mailing
list. It's pretty moribund compared to around here.

I am primarily interested in correct code (be it from a specification,
or not), and in looking at packet captures, to validate that it is
doing what it says on the tin, and moreover getting more stuff on
those tin, that I already know, should be tested for.

I agree that someday trying to nail down what latency under load
means, would be good to do. I'd settle at the moment, for single flow
simultaneous, tcp up, down, ping, and within stream latencies... all
plotted on the same chart.

> JL
>


-- 
This song goes out to all the folk that thought Stadia would work:
https://www.linkedin.com/posts/dtaht_the-mushroom-song-activity-698135607352320-FXtz
Dave Täht CEO, TekLibre, LLC
___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat


Re: [Bloat] [EXTERNAL] Re: [Starlink] Researchers Seeking Probe Volunteers in USA

2023-01-09 Thread Livingood, Jason via Bloat
> 0) None of the tests last long enough.

The user-initiated ones tend to be shorter - likely because the average user 
does not want to wait several minutes for a test to complete. But IMO this is 
where a test platform like SamKnows, Ookla's embedded client, NetMicroscope, 
and others can come in - since they run in the background on some randomized 
schedule w/o user intervention. Thus, the user's time-sensitivity is no longer 
a factor and a longer duration test can be performed.

> 1) Not testing up + down + ping at the same time

You should consider publishing a LUL BCP I-D in the IRTF/IETF - like in IPPM...

JL

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