Flent-users discussion archives
 help / color / mirror / Atom feed
From: Dave Taht <dave.taht@gmail.com>
To: Sebastian Moeller <moeller0@gmx.de>
Cc: flent-users <flent-users@flent.org>
Subject: [Flent-users] Re: flaws and features of flent
Date: Mon, 16 Jan 2023 08:28:17 -0800	[thread overview]
Message-ID: <CAA93jw4GPzLz3orkQXLi9pHL11RzJGT8MS1X1hsBUeKfjUvzPQ@mail.gmail.com> (raw)
In-Reply-To: <1D4C239D-C890-4FFA-9256-090D964F36C0@gmx.de>

On Sun, Jan 15, 2023 at 11:14 PM Sebastian Moeller <moeller0@gmx.de> wrote:
>
> Hi Dave,
>
> On 16 January 2023 02:52:45 CET, Dave Taht <dave.taht@gmail.com> wrote:
> >A rather long, meandering piece:
> >
> >https://blog.cerowrt.org/post/flaws_in_flent/
>
> I think 'Flent’s default sampling rate is too low' has a conceptually simple solution, record dats at high sample rates and optionally downsample before plotting... However with the range of network speeds flent covers picking a sane default is going to be a challenge. With almost all endhosts sporting gigabit ethernet interfaces and beyond I am unsure whether basing the sample rate on to-be-detected interface rate would help (it might serve to give an upper bound for sampling rate).

The real problem is we run out of cpu when polling ss too fast. I
haven't played with google labs' sampler as yet, it's here:

A righter answer, long term, is to start wedging a lag meter, and
sampling TCP_INFO, directly into more utilities and libraries. Adding
a lag meter to git, for example, would get some eyeballs on
bufferbloat quickly. Something that worked within a web browser,
passively, would be good also.

Getting TCP_INFO sampling into rust and go has been on my mind.

https://github.com/rust-lang/socket2/issues/362

And thank you for reading. I am loving having emacs in front of me
again. I hope to move more of my workload back into that, including
email, and to add a grammar checker, in the hope that more of what I
write in the future is more accessible.

Assuming I can keep pounding out text at this rate, and continue to
improve my workflow, the bufferbloat.net site will improve rapidly.

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



-- 
This song goes out to all the folk that thought Stadia would work:
https://www.linkedin.com/posts/dtaht_the-mushroom-song-activity-6981366665607352320-FXtz
Dave Täht CEO, TekLibre, LLC

  reply	other threads:[~2023-01-16 16:28 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-01-16  1:52 [Flent-users] " Dave Taht
2023-01-16  7:14 ` [Flent-users] " Sebastian Moeller
2023-01-16 16:28   ` Dave Taht [this message]
2023-01-16  7:24 ` Sebastian Moeller
2023-01-16 16:30   ` Dave Taht

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.flent.org/postorius/lists/flent-users.flent.org/

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

  git send-email \
    --in-reply-to=CAA93jw4GPzLz3orkQXLi9pHL11RzJGT8MS1X1hsBUeKfjUvzPQ@mail.gmail.com \
    --to=dave.taht@gmail.com \
    --cc=flent-users@flent.org \
    --cc=moeller0@gmx.de \
    --subject='[Flent-users] Re: flaws and features of flent' \
    /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

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox