[Cerowrt-devel] collecting some data?

[Cerowrt-devel] collecting some data?

[Dave Taht]

[-- Attachment #1.1: Type: text/plain, Size: 2108 bytes --]

I have been collecting statistics on the behavior of the 3.7.5-2 version of
cerowrt, and perhaps y'all out there can help. I've been doing four
repeatable tests.

I feed the attached file into the chrome web page benchmark and run it 4
times:

1) No other load on the system
2) while running a single up and single down netperf stream to icei.org (on
the east coast)
3) then while still loaded, after turning on simple_qos.sh, set for ~85% of
the rated up/down bandwidth...
4) then killing the load, and keeping simple_qos enabled.

I export each detailed (you have to select it) .csv output from that
benchmark to a file.

I do it against a bidirectional stream - one each of

netperf -l 6000 -4 -H icei.org -t TCP_MAERTS &
netperf -l 6000 -4 -H icei.org -t TCP_STREAM &

(if you have ipv6, more power to you, kill the -4. )

then starting the web tests 10-20 seconds later. It's interesting to watch
cero's bandwidth graphs while doing this.

There are netperf servers setup on the east (icei.org) and west coast (
snapon.lab.bufferbloat.net) and a few other places. It would be sane to
setup your own someplace you control (from netperf's svn compiled with
--enable-demo), as it helps to have a shorter RTT to truly load up the
link. You can also run the rrul test (preferably for 5 minutes or more
because this test idea KILLs web page load times) to get a heavier load and
more graphs.

If you are up to trying this test series, please let me know, send along
the details of your setup, and the 4 csv files, and perhaps your data will
show up in an upcoming paper.

The chrome web page benchmark can be had at:

https://chrome.google.com/webstore/detail/page-benchmarker/channimfdomahekjcahlbpccbgaopjll?hl=en

And requires you fire off chrome with --enable-benchmarking

You can use the version of netperf on cerowrt to generate the load if you
like, but rrul doesn't run on that. Latest version of rrul is at
https://github.com/tohojo/netperf-wrapper

-- 
Dave Täht

Fixing bufferbloat with cerowrt:
http://www.teklibre.com/cerowrt/subscribe.html

[-- Attachment #1.2: Type: text/html, Size: 2898 bytes --]

[-- Attachment #2: 12.urls --]
[-- Type: application/octet-stream, Size: 302 bytes --]

http://www.google.com,
https://www.facebook.com,
http://www.youtube.com,
http://www.yahoo.com,
http://www.baidu.com,
http://www.wikipedia.org,
http://www.amazon.com,
https://twitter.com,
http://www.bing.com,
http://www.yandex.ru,
http://www.microsoft.com/en-US/default.aspx,
http://xfinity.comcast.net

Re: [Cerowrt-devel] collecting some data?

[Dave Taht]

[-- Attachment #1: Type: text/plain, Size: 2512 bytes --]

to shorten the painful duration of the test, you can tell the chrome
benchmark merely to repeat 3 times each rather than 10.

On Mon, Feb 25, 2013 at 3:13 PM, Dave Taht <dave.taht@gmail.com> wrote:

> I have been collecting statistics on the behavior of the 3.7.5-2 version
> of cerowrt, and perhaps y'all out there can help. I've been doing four
> repeatable tests.
>
> I feed the attached file into the chrome web page benchmark and run it 4
> times:
>
> 1) No other load on the system
> 2) while running a single up and single down netperf stream to icei.org(on the east coast)
> 3) then while still loaded, after turning on simple_qos.sh, set for ~85%
> of the rated up/down bandwidth...
> 4) then killing the load, and keeping simple_qos enabled.
>
> I export each detailed (you have to select it) .csv output from that
> benchmark to a file.
>
> I do it against a bidirectional stream - one each of
>
> netperf -l 6000 -4 -H icei.org -t TCP_MAERTS &
> netperf -l 6000 -4 -H icei.org -t TCP_STREAM &
>
> (if you have ipv6, more power to you, kill the -4. )
>
> then starting the web tests 10-20 seconds later. It's interesting to watch
> cero's bandwidth graphs while doing this.
>
> There are netperf servers setup on the east (icei.org) and west coast (
> snapon.lab.bufferbloat.net) and a few other places. It would be sane to
> setup your own someplace you control (from netperf's svn compiled with
> --enable-demo), as it helps to have a shorter RTT to truly load up the
> link. You can also run the rrul test (preferably for 5 minutes or more
> because this test idea KILLs web page load times) to get a heavier load and
> more graphs.
>
> If you are up to trying this test series, please let me know, send along
> the details of your setup, and the 4 csv files, and perhaps your data will
> show up in an upcoming paper.
>
> The chrome web page benchmark can be had at:
>
>
> https://chrome.google.com/webstore/detail/page-benchmarker/channimfdomahekjcahlbpccbgaopjll?hl=en
>
> And requires you fire off chrome with --enable-benchmarking
>
> You can use the version of netperf on cerowrt to generate the load if you
> like, but rrul doesn't run on that. Latest version of rrul is at
> https://github.com/tohojo/netperf-wrapper
>
> --
> Dave Täht
>
> Fixing bufferbloat with cerowrt:
> http://www.teklibre.com/cerowrt/subscribe.html
>



-- 
Dave Täht

Fixing bufferbloat with cerowrt:
http://www.teklibre.com/cerowrt/subscribe.html

[-- Attachment #2: Type: text/html, Size: 3742 bytes --]

Re: [Cerowrt-devel] collecting some data?

[Maciej Soltysiak]

[-- Attachment #1: Type: text/plain, Size: 768 bytes --]

On Mon, Feb 25, 2013 at 9:13 PM, Dave Taht <dave.taht@gmail.com> wrote:

> I have been collecting statistics on the behavior of the 3.7.5-2 version
> of cerowrt, and perhaps y'all out there can help. I've been doing four
> repeatable tests.
>
> I feed the attached file into the chrome web page benchmark and run it 4
> times:
>
> 1) No other load on the system
> 2) while running a single up and single down netperf stream to icei.org(on the east coast)
> 3) then while still loaded, after turning on simple_qos.sh, set for ~85%
> of the rated up/down bandwidth...
> 4) then killing the load, and keeping simple_qos enabled.
>
Is that good if I run the chrome benchmark from a win7 laptop on 802.11n
wifi, while netperfs are going from the wndr3800?

Regards,
Maciej

[-- Attachment #2: Type: text/html, Size: 1186 bytes --]

Re: [Cerowrt-devel] collecting some data?

[Maciej Soltysiak]

[-- Attachment #1: Type: text/plain, Size: 1405 bytes --]

On Mon, Feb 25, 2013 at 9:39 PM, Maciej Soltysiak <maciej@soltysiak.com>wrote:

> On Mon, Feb 25, 2013 at 9:13 PM, Dave Taht <dave.taht@gmail.com> wrote:
>
>> I have been collecting statistics on the behavior of the 3.7.5-2 version
>> of cerowrt, and perhaps y'all out there can help. I've been doing four
>> repeatable tests.
>>
>> I feed the attached file into the chrome web page benchmark and run it 4
>> times:
>>
>> 1) No other load on the system
>> 2) while running a single up and single down netperf stream to icei.org(on the east coast)
>> 3) then while still loaded, after turning on simple_qos.sh, set for ~85%
>> of the rated up/down bandwidth...
>> 4) then killing the load, and keeping simple_qos enabled.
>>
> Is that good if I run the chrome benchmark from a win7 laptop on 802.11n
> wifi, while netperfs are going from the wndr3800?
>
Forgot to mention my win7 TCP setup is:
netsh interface>tcp show global
Querying active state...

TCP Global Parameters
----------------------------------------------
Receive-Side Scaling State          : enabled
Chimney Offload State               : automatic
NetDMA State                        : enabled
Direct Cache Acess (DCA)            : disabled
Receive Window Auto-Tuning Level    : normal
Add-On Congestion Control Provider  : none
ECN Capability                      : enabled
RFC 1323 Timestamps                 : disabled

Regards,
Maciej

[-- Attachment #2: Type: text/html, Size: 2296 bytes --]

Re: [Cerowrt-devel] collecting some data?

[Dave Taht]

[-- Attachment #1: Type: text/plain, Size: 1201 bytes --]

On Mon, Feb 25, 2013 at 3:39 PM, Maciej Soltysiak <maciej@soltysiak.com>wrote:

> On Mon, Feb 25, 2013 at 9:13 PM, Dave Taht <dave.taht@gmail.com> wrote:
>
>> I have been collecting statistics on the behavior of the 3.7.5-2 version
>> of cerowrt, and perhaps y'all out there can help. I've been doing four
>> repeatable tests.
>>
>> I feed the attached file into the chrome web page benchmark and run it 4
>> times:
>>
>> 1) No other load on the system
>> 2) while running a single up and single down netperf stream to icei.org(on the east coast)
>> 3) then while still loaded, after turning on simple_qos.sh, set for ~85%
>> of the rated up/down bandwidth...
>> 4) then killing the load, and keeping simple_qos enabled.
>>
> Is that good if I run the chrome benchmark from a win7 laptop on 802.11n
> wifi, while netperfs are going from the wndr3800?
>

sure.

So long as the circumstances are documented, all data is good. (well, a
start!) I note it would be helpful to also collect the ping rtt to the
netperf server you are using....


>
> Regards,
> Maciej
>



-- 
Dave Täht

Fixing bufferbloat with cerowrt:
http://www.teklibre.com/cerowrt/subscribe.html

[-- Attachment #2: Type: text/html, Size: 2186 bytes --]