Hi all, >>> 2) CPU share enforcement test with kernbench >>> >>> User and Sys time are almost constant in every case. >>> On the other hand, enforcing shares are _roughly_ equal to >>> the percentages of CPU consumed by kernbench. It is not too bad, >>> but further investigation is needed. >>> >>> Share[%] Elaps[s] User[s] Sys[s] CPU[%] >>> ---------------------------------------------- >>> 100 321.9 295.1 19.7 97.0 >>> 90 325.4 293.5 19.5 95.8 >>> 80 345.9 293.2 19.7 90.0 >>> 70 392.6 293.3 19.6 79.2 >>> 60 466.0 293.6 19.8 66.8 >>> 50 583.0 296.4 19.8 53.8 >>> 40 725.3 297.7 19.6 43.2 >>> 30 925.2 299.6 19.4 34.0 >>> 20 1458.2 301.5 19.3 21.2 >>> 10 2595.5 303.4 20.0 12.0 >>> ---------------------------------------------- >>> >>> - One class for running kernbench was defined and various >>> share was set to the class. >>> >>> - kernbench -M against 2.6.15 kernel source >>> >>> - Five infinity loop programs were running on the default_class >>> at the same time. Otherwise, kernbench was able to consume >>> all of the CPU resource. >> Nice results! >> >> Out of curiosity, do you know why 1 infinite loop was not sufficient as >> a load? > > Because these measurements were done by a single CPU system, 1 infinite > loop was sufficient. But it is not true for multi CPU systems.
I am also curious to know whether 1 infinite loop is really sufficient or not. I reran kernbench on the same machine with 1 infinite loop. The following is the results. Share[%] Elaps[s] User[s] Sys[s] CPU[%] -------------------------------------------- 100 321.7 295.8 20.0 97.7 90 329.2 295.7 19.8 95.4 80 349.9 295.1 19.4 89.4 70 382.0 295.2 19.8 81.8 60 498.4 296.2 19.6 67.0 50 561.9 298.3 19.5 56.0 40 684.3 300.7 19.9 46.0 30 846.9 301.6 19.6 37.6 20 1309.6 305.0 19.7 24.0 10 2209.0 306.7 20.0 14.0 -------------------------------------------- The percentage of CPU kernbench consumed is still _roughly_ equal to the enforcing share. However, a tendency to overshoot the percentage of CPU is rather emphasized than the 5 infinite loops case. 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
