[Bloat] [Rpm] An 8 years perspective on broadband in the usa - fcc data

[Sam Crawford]

Thanks for the heads up.

Downstream throughput and upstream throughput are measured in indepdent
tests, they aren't mixed as the question suggests might be the case. So,
latency under downstream load is measured independently from latency under
upstream load. The raw data for both is published for the FCC project.

I've not read the paper to understand which one they are using. But I'll
take a look tomorrow.

Thanks

Sam


On Sun, 19 Jun 2022, 20:39 Matt Mathis, <matt.mathis at gmail.com> wrote:

> Sam, see the question below about your methodology, from your FCC report.
>
> Thanks,
> --MM--
> Evil is defined by mortals who think they know "The Truth" and use force
> to apply it to others.
> -------------------------------------------
> Matt Mathis  (Email is best)
> Home & mobile: 412-654-7529 please leave a message if you must call.
>
>
>
> On Sun, Jun 19, 2022 at 8:46 AM Sebastian Moeller via Rpm <
> rpm at lists.bufferbloat.net> wrote:
>
>> Hi Dave,
>>
>> > On Jun 19, 2022, at 17:32, Dave Taht via Bloat <
>> bloat at lists.bufferbloat.net> wrote:
>> >
>> > It's not clear if they are only measuring downloads in fig 4. ?
>> >
>> >
>> https://vaibhavbajpai.com/documents/papers/proceedings/fcc-networking-2022.pdf
>>
>> indeed:
>>
>> "The SamKnows probes measure a variety of broadband met- rics [22] over
>> IPv4. The tests [22] covered in this paper include: multi-threaded download
>> speed, UDP latency, UDP latency un- der load, and UDP packet loss. The
>> tests connect to the closest measurement servers (based on round-trip
>> time); measurement servers are hosted both off-net (outside of ISP
>> boundary) and on-net (within ISP boundary). The off-net measurement points
>> were hosted by Measurement Lab (M-Lab) [23] and changed to Level 3
>> Communications (CenturyLink) across ten cities in the US, while the on-net
>> measurement targets are hosted by the ISPs themselves. The FCC publishes
>> the raw data for each month [24], along with annual reports on major
>> findings [25]. For controllability reasons, the annual MBA reports only
>> cover measurements to the off-net locations, i.e., latency results might be
>> inflated due to geographical distance. Thus, we also consider the on-net
>> measurements in our analysis."
>>
>> reference 22 points to
>> https://www.fcc.gov/general/measuring-broadband-america-open-methodology
>> "Latency under load     Average round trip time for a series of regularly
>> spaced UDP packets sent during downstream/upstream sustained tests"
>>
>> this seems to imply that it would be mixed upload/download saturation?
>>
>>
>> But I agree that this is unclear... maybe Sam (Crawford) knows (pun
>> intended), I do not have his email, but maybe he is on-list or someone else
>> can relay that question?
>>
>> Regards
>>         Sebastian
>>
>>
>> >
>> >
>> > --
>> > FQ World Domination pending:
>> https://blog.cerowrt.org/post/state_of_fq_codel/
>> > Dave Täht CEO, TekLibre, LLC
>> > _______________________________________________
>> > Bloat mailing list
>> > Bloat at lists.bufferbloat.net
>> > https://lists.bufferbloat.net/listinfo/bloat
>>
>> _______________________________________________
>> Rpm mailing list
>> Rpm at lists.bufferbloat.net
>> https://lists.bufferbloat.net/listinfo/rpm
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.bufferbloat.net/pipermail/bloat/attachments/20220619/9b116a83/attachment-0001.html>