From: Dave Taht <dave.taht@gmail.com>
To: Federico Capoano <f.capoano@openwisp.io>,
Battle of the Mesh Mailing List <battlemesh@ml.ninux.org>,
Make-Wifi-fast <make-wifi-fast@lists.bufferbloat.net>,
libreqos <libreqos@lists.bufferbloat.net>
Subject: Re: [Make-wifi-fast] [Battlemesh] Testbed v15 (2023)
Date: Mon, 20 Mar 2023 09:34:34 -0700 [thread overview]
Message-ID: <CAA93jw6bv6OY43mPBV-nwe3JLqa44-JzAcxHaXjLQ_tLesvmVQ@mail.gmail.com> (raw)
In-Reply-To: <CAAGgX6JX9+MK2puihohJ2OEwomNVB+bugA0k9jGv2cXVPt_mVw@mail.gmail.com>
I am not planning to come to battlemesh once again. :( However, I
would love to steer y'all at the wonderful new passive monitoring
tools that kathie nichols (pping), simon sundberg and toke and jesper
(epping) and more recently the libreqos.io project has developed to
where they can run at 10s of gbits per core and gain unprecedented
insight into the real behaviors of real flows, sampling at 10ms
interval. Slow start, sawtooths, the impact of one flow into another,
and underlying wifi behaviors are all observable.
I do hope you use flent to drive some tests! I am so hopeful for
wonderful improvements in the rtt_fair test series as well, and I am
dying to hear news of how well the latest babeld is performing.
For a live demo of libreqos 1.5 (heimdall branch), please click on
https://payne.taht.net and hit "bandwidth test" if one is not already
running. It is flent driving the tests (as well as broadband forum's
tr-471), one of my favorites is now the tcp_4up_squarewave test
because I love seeing the waveforms crash into each other.
Libreqos basically leverages ebpf, XDP, cake diffserv4, and a ton of
rust to do its magic, it can be configured to transparently bridge a
network, and even work to bridge across two vlans. It takes about 20
minutes to compile, but we do have a .deb available for the upcoming
1.4 release which cuts the install time down to 5 minutes. It is too
big and too complex to run on openwrt (maybe next year?), and thus we
use ubuntu for it, and have it pushing well over 10k subscribers
through cake for multiple ISPs now at greater than 13Gbits.
We gave a talk about it over here:
https://packetpushers.net/podcast/heavy-networking-666-improving-quality-of-experience-with-libreqos/
github here: https://github.com/LibreQoE/LibreQoS
parent reply other threads:[~2023-03-20 16:34 UTC|newest]
Thread overview: expand[flat|nested] mbox.gz Atom feed
[parent not found: <CAAGgX6JX9+MK2puihohJ2OEwomNVB+bugA0k9jGv2cXVPt_mVw@mail.gmail.com>]
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/make-wifi-fast.lists.bufferbloat.net/
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=CAA93jw6bv6OY43mPBV-nwe3JLqa44-JzAcxHaXjLQ_tLesvmVQ@mail.gmail.com \
--to=dave.taht@gmail.com \
--cc=battlemesh@ml.ninux.org \
--cc=f.capoano@openwisp.io \
--cc=libreqos@lists.bufferbloat.net \
--cc=make-wifi-fast@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