-----Original Message-----
From: MAEDA Naoaki [mailto:[EMAIL PROTECTED]
Sent: Thu 2/16/2006 1:00 PM
To: Suryanarayanan, Rajaram
Cc: [email protected]
Subject: Re: [ckrm-tech] How to see the benefits of CKRM ?
Hi,
> In my case, I have a single CPU machine. Is there any tests already, to test
> CPU resource controller and feel the difference of using CKRM ? or to
> demonstrate CKRM ?
> Any suggestions ?
I've uploaded the follwing two test results to demonstrate the CPU resource
controller for f-series CKRM.
http://prdownloads.sourceforge.net/ckrm/cpurc-v0.3-2615-kernbench.pdf?download
http://prdownloads.sourceforge.net/ckrm/cpurc-v0.3-2615-lat_ctx.pdf?download
The former demonstrates that the share is roughly proportional to the CPU usage
of kernbench. The latter demonstrates that the CPU controller's overhead is
very small by comparing the scheduling overhead of vannila 2.6.15 to
2.6.15 with CPU controller using lat_ctx.
I've done these tests manually, but if you want, you can reproduce
the results.
Thanks,
MAEDA Naoaki
Thanks again for your prompt reply.
But I am sorry. I have no knowledge on kernbench and lat_ctx etc and yet to
look in to what they are...
I guess that you usually use kernbench to see CPU resource controller working.
I was actually wondering whether there is any simple way like creating two
processes, then allocating them to different classes and then see that one runs
faster than another.
btw..what is vannila 2.6.15 ? Sorry for my ignorance.
Thanks,
Rajaram.
-------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc. Do you grep through log files
for problems? Stop! Download the new AJAX search engine that makes
searching your log files as easy as surfing the web. DOWNLOAD SPLUNK!
http://sel.as-us.falkag.net/sel?cmd=lnk&kid3432&bid#0486&dat1642
_______________________________________________
ckrm-tech mailing list
https://lists.sourceforge.net/lists/listinfo/ckrm-tech
