Hi all, MAEDA Naoaki wrote: > Hi Matt, > > Thank you for your comments. > > Matt Helsley wrote: >> On Tue, 2006-01-24 at 18:25 +0900, MAEDA Naoaki wrote: >>> Hi all, >>> >>> I've taken some benchmark results of cpurc-v0.3-2615 in order to >>> measure its overhead and the accuracy of enforced CPU share. >>> >>> 1) Overhead measurement with lat_ctx -s 0 in various # of processes >>> >>> The overhead is between -0.02 to +0.07 microseconds. >>> It is quite small. >>> >>> # Vanilla 2.6.15 2.6.15+cpurc-v0.3 Diff [us] >>> --------------------------------------------------------- >>> 2 0.51 0.52 +0.01 >>> 4 0.58 0.6 +0.02 >>> 8 0.71 0.71 0.0 >>> 16 0.72 0.72 0.0 >>> 32 0.73 0.72 -0.01 >>> 64 0.78 0.85 +0.07 >>> 128 1.08 1.06 -0.02 >>> 256 1.85 1.89 +0.04 >>> ---------------------------------------------------------- >> These numbers look pretty good. Are you doing multiple runs of lat_ctx >> -s 0 at the same number of processes in order to get some idea of how >> noisy the difference is? Perhaps these values are within the noise >> margin. > > The results were average of three runs of "lat_ctx -s 0 -N 1000". > Probably these values are within the noise margin.
I've uploaded a line chart that shows more broad range of lat_ctx to the project web site. (The size is not suitable for being attached to e-mail.) http://prdownloads.sourceforge.net/ckrm/cpurc-v0.3-2615-lat_ctx.pdf?download According to the chart, the differences between 2.6.15 and cpurc-v0.3 seem to be within the noise margin. Thanks, MAEDA Naoaki ------------------------------------------------------- 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&kid=103432&bid=230486&dat=121642 _______________________________________________ ckrm-tech mailing list https://lists.sourceforge.net/lists/listinfo/ckrm-tech
