From: Henry N. <Hen...@Ar...> - 2008-03-22 18:51:13
|
Óscar Fuentes wrote: > Henry Nestler <Hen...@Ar...> writes: > >>> The test suite starts a process more than 1000 times, pipes the output >>> into the controlling process, etc. My experience shows that GNU/Linux is >>> more effective than Windows for this sort of tasks. >> Suggest from you, here is a test script, that only eceute some >> processes, without disk io: >> #!/bin/bash >> >> loop=10000 >> while test $loop -gt 0 >> do >> x=`uname -r` >> loop=$(($loop - 1)) >> done >> >> Tested with Intel(R) Pentium(R) 4 Mobile CPU 1.80GHz >> 22.9s native, kernel 2.6.17 >> 89.7s coLinux 0.8.0, kernel 2.6.22.18 >> >> Can you run tis on WVware please? > > 26.2 seconds > Pentium M 2.00 GHz > VMWare 6.0.0 > Linux etch 2.6.18-5-686 #1 SMP Ok. And is not faster as my native. An other PC: I have 17.1s on Intel(R) Core(TM)2 CPU 6400 @ 2.13GHz. Runs only on one core and with 1.6GHz (2.6.22.5-31 SuSE 10.3 native) > I can't test it on coLinux, because I corrupted fstab while trying to > mount with noatime: I made a typo (noattime) and now the file system > mounts read-only. umount does not work, nor mount -o remount, etc. > By the while "noatime" does not improve speed. I have tested with and without. Differ coLinux version made no effect. All test on same machine, with SuSE 9.0, coLinux runs on partitions (dualboot). make-3.81.tar.gz unpack config make Native SuSE90 1.8s 28.3s 14.6s kernel 2.6.17 colinux0.7.3 1.7s 63.6s 28.2s (first, after reboot Windows) 1.0s 64.4s 28.6s 0.6s 61.1s 24.4s 0.9s 64.2s 28.6s 1.0s 64.0s 29.1s colinux0.7.2 0.9s 54.3s 28.6s colinux0.8.0 0.9s 64.9s 28.7s (noatime) 0.8s 65.7s 28.7s (atime) colinux080async 0.8s 65.4s 29.0s 0.9s 65.8s 29.4s 0.7s 62.2s 28.9s colinux0.6.4 0.8s 64.6s 27.7s 0.8s 61.2s 29.0s 0.8s 63.4s 28.8s coLinux speed is 50% slower as native. -- Henry N. |