From: Dave Taht <dave.taht@gmail.com>
To: "cerowrt-devel@lists.bufferbloat.net"
<cerowrt-devel@lists.bufferbloat.net>
Subject: [Cerowrt-devel] pie testing
Date: Tue, 25 Mar 2014 12:13:25 -0700 [thread overview]
Message-ID: <CAA93jw6JrcWcsfk-vyCwqHsQ7YNcdjdM+9dO2OGk6CRyiaVSxw@mail.gmail.com> (raw)
I have been really kind about waiting (nearly 2 years) for pie to
stabilize and the
final version to hit linux mainline. Admittedly the version as planned to
be implemented in cable modems is so very different as for me
to despair of getting comparable results, but...
Now that it is apparently stable, and cero seems stable...
I'd appreciate some results using it
(with the default pie target of 20ms and with a target of 5ms) with tests like
rich's new stuff, and rrul... I'm particularly interested in DSL results.
pie was extensively simulated using ns2 models that I disagree with on
many points, but does appear to work fairly well for simple models of traffic,
which was what it was mostly tested against. On my todo list is to try
combining drr + pie or qfq + pie.
pie target 20ms:
root@comcast-gw:~# ./speedtest.sh
Testing against netperf.richb-hanover.com while pinging gstatic.com
(60 seconds in each direction)
............................................................
Download: 24.6 Mbps
Latency: (in msec, 61 pings, 0.00% packet loss)
Min: 15.719
10pct: 18.012
Median: 22.916
Avg: 26.091
90pct: 41.447
Max: 55.721
.............................................................
Upload: 3.95 Mbps
Latency: (in msec, 61 pings, 0.00% packet loss)
Min: 17.539
10pct: 22.938
Median: 31.679
Avg: 36.764
90pct: 45.287
Max: 145.407
root@comcast-gw:~# ./speedtest.sh -H snapon.lab.bufferbloat.net
Testing against snapon.lab.bufferbloat.net while pinging gstatic.com
(60 seconds in each direction)
............................................................
Download: 27.61 Mbps
Latency: (in msec, 61 pings, 0.00% packet loss)
Min: 16.082
10pct: 22.831
Median: 30.295
Avg: 29.515
90pct: 35.582
Max: 39.558
............................................................
Upload: 4.13 Mbps
Latency: (in msec, 61 pings, 0.00% packet loss)
Min: 16.697
10pct: 34.470
Median: 44.720
Avg: 45.103
90pct: 54.880
Max: 62.281
For direct comparison (not that I recommend a 20ms target for fq_codel!
pie and codel treat the target idea very differently! )
fq_codel target 20ms
root@comcast-gw:~# ./speedtest.sh -H snapon.lab.bufferbloat.net
Testing against snapon.lab.bufferbloat.net while pinging gstatic.com
(60 seconds in each direction)
............................................................
Download: 27.97 Mbps
Latency: (in msec, 61 pings, 0.00% packet loss)
Min: 17.132
10pct: 18.177
Median: 21.482
Avg: 21.786
90pct: 24.725
Max: 37.424
............................................................
Upload: 4.13 Mbps
Latency: (in msec, 61 pings, 0.00% packet loss)
Min: 18.813
10pct: 21.308
Median: 24.863
Avg: 24.777
90pct: 27.931
Max: 29.007
rrul test results for both of these are at:
http://snapon.lab.bufferbloat.net/~d/cerowrt-3.10.32-12/pie
--
Dave Täht
Fixing bufferbloat with cerowrt: http://www.teklibre.com/cerowrt/subscribe.html
reply other threads:[~2014-03-25 19:13 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/cerowrt-devel.lists.bufferbloat.net/
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=CAA93jw6JrcWcsfk-vyCwqHsQ7YNcdjdM+9dO2OGk6CRyiaVSxw@mail.gmail.com \
--to=dave.taht@gmail.com \
--cc=cerowrt-devel@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