Cake - FQ_codel the next generation
 help / color / mirror / Atom feed
From: Dave Taht <dave.taht@gmail.com>
To: freeswitch-dev@lists.freeswitch.org,
	freeswitch-users@lists.freeswitch.org,
	cake@lists.bufferbloat.net
Subject: [Cake] testers wanted for low latency voip/video conferencing qdisc
Date: Sun, 18 Oct 2015 13:24:18 +0200	[thread overview]
Message-ID: <CAA93jw7xEwgeOweAU5fyRnV-RyfN8YoV0iZ613Z-CpoJvMTXzw@mail.gmail.com> (raw)

Dear Freeswitch folk:

It is generally my hope that many of you have been using things like
fq_codel (now the default in fedora, and many other linux derived
OSes) and qos systems like the "sqm"[1] system now in openwrt, and
streamboost, etc in order to improve voip and videoconferencing
behavior along the edge, and beat bufferbloat.

For the past 3 years we have been working on a successor to fq_codel,
called "cake" - which among many other improvements like 8way set
associative hashing, and less cpu use, also obeys what we know of the
defined diffserv markings for voip and video conferencing - not that
we think they are widely used as yet.

We are finally at the point where cake can be tested more widely with
an out of tree kernel module, and patches to tc. A complete technical
description of cake, links the github repos, and how to install are
at:

http://www.bufferbloat.net/projects/codel/wiki/CakeTechnical

And we hope to submit to kernel mainline after more testing gets done.

We have many tests (see flent.org for the flent tool) for voip but we
have yet to develop good ones for videoconferencing, and in addition
to any testing of cake you can do, I would love to know how we too can
add tests for that to our test suites.

I had a chance last friday to come on the vuc show and talk about
fq_codel, cake, bloat, etc on the latter half, and the first half was
dominated by our controversial proposal to fix the edge of the
internet. - but my heart is in seeing the technical fixes for better
voip and videoconferencing get more widely used so we can finally
deploy them!

https://www.youtube.com/watch?v=yGYZ83nq4Js
http://www.businesswire.com/news/home/20151014005564/en (for the
letter we were talking about)

And my group of bufferbloat folk plan to be on the cluecon show nov 18th.

More freeswitch data to toss around - logs - packet captures -
empirical evidence - anything would be great to have before then. TIA.

[1] http://wiki.openwrt.org/doc/howto/sqm

-- 
Dave Täht
Do you want faster, better, wifi? https://www.patreon.com/dtaht

                 reply	other threads:[~2015-10-18 11:24 UTC|newest]

Thread overview: [no followups] expand[flat|nested]  mbox.gz  Atom feed

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/cake.lists.bufferbloat.net/

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

  git send-email \
    --in-reply-to=CAA93jw7xEwgeOweAU5fyRnV-RyfN8YoV0iZ613Z-CpoJvMTXzw@mail.gmail.com \
    --to=dave.taht@gmail.com \
    --cc=cake@lists.bufferbloat.net \
    --cc=freeswitch-dev@lists.freeswitch.org \
    --cc=freeswitch-users@lists.freeswitch.org \
    /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