General list for discussing Bufferbloat
 help / color / mirror / Atom feed
From: Sebastian Moeller <moeller0@gmx.de>
To: Dave Taht <dave.taht@gmail.com>,
	Dave Taht via Bloat <bloat@lists.bufferbloat.net>,
	Rpm <rpm@lists.bufferbloat.net>,
	libreqos <libreqos@lists.bufferbloat.net>,
	bloat <bloat@lists.bufferbloat.net>
Subject: Re: [Bloat] cloudflare on a roll
Date: Wed, 19 Apr 2023 08:36:20 +0200	[thread overview]
Message-ID: <0988FDEA-65D4-471D-9CE4-8A832013BF13@gmx.de> (raw)
In-Reply-To: <CAA93jw4jAZ=41mRUtuqzHEquWYwsts8hvUKm81s2dP3TmehS8Q@mail.gmail.com>

[-- Attachment #1: Type: text/plain, Size: 2467 bytes --]

Also this:
https://blog.cloudflare.com/aim-database-for-internet-quality/

A pretty decent article explaining the issue in a very accessible way.

Nitpicks (in case David Tuber is on this list):
A) jitter is important for gaming (if only as it affects the necessary added delay for de-jitter buffers), just think extreme cases like packet reordering, so I think the gaing score should include the jitte sub-score.
B) "You have to make video calls to her clients all day and you sit in the office farthest away from the wireless access point." Probably "calls to your clients" instead.

Also I am a big fan of the cloudflare test including a packetloss test under idle conditions. This was really helpful in getting enough users to chip in loss data to convince my ISP that they had a severe loss problem (sustained packetloss from ~1-10% even under idle conditions like 1AM), which in turn resulted in ISP and some not revealed manufacturer going ona deep debug session and apparently squelching the issue.
Sidenotes.:
A) even at 0.8% random loss TCPs struggle to keep utilisation high, and it takes an ungodly numer of concurrent flows to saturate even a moderate link.
B) tying into an earlier discussion about monitoring latency/loss along network paths and the utility of doing so for endusers, I caught my ISPs attention by posting a series of MTR traces taken at the middle of the night against a set of DNS servers (including my ISP's). All showing increased sustained loss after a single gateway in Hamburg (and no such loss for other gateways of the same ISP). There had been complaints of reduced download rates before, but it apparently it was the longitudinal set of traces that made their NOC take the matter seriously and spring into action. Yes, that ISP should have found/fixed the issue pro-actively, but to paraphrase D. Rumsfeld, you go on the internet with the ISP you have, not the one you want to have....

Regards
        Sebastian


On 18 April 2023 17:31:44 CEST, Dave Taht via Bloat <bloat@lists.bufferbloat.net> wrote:
>https://blog.cloudflare.com/making-home-internet-faster/
>
>
>-- 
>AMA March 31: https://www.broadband.io/c/broadband-grant-events/dave-taht
>Dave Täht CEO, TekLibre, LLC
>_______________________________________________
>Bloat mailing list
>Bloat@lists.bufferbloat.net
>https://lists.bufferbloat.net/listinfo/bloat

-- 
Sent from my Android device with K-9 Mail. Please excuse my brevity.

[-- Attachment #2: Type: text/html, Size: 2767 bytes --]

      parent reply	other threads:[~2023-04-19  6:36 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-04-18 15:31 Dave Taht
2023-04-18 16:31 ` Dave Collier-Brown
2023-04-18 18:05 ` [Bloat] [Rpm] " rjmcmahon
2023-04-19  6:36 ` Sebastian Moeller [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

  List information: https://lists.bufferbloat.net/postorius/lists/bloat.lists.bufferbloat.net/

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=0988FDEA-65D4-471D-9CE4-8A832013BF13@gmx.de \
    --to=moeller0@gmx.de \
    --cc=bloat@lists.bufferbloat.net \
    --cc=dave.taht@gmail.com \
    --cc=libreqos@lists.bufferbloat.net \
    --cc=rpm@lists.bufferbloat.net \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox