R: performance gap between 2.2.14 and 2.4.0-test. kernels------------Here we are with kernel 2.4.0-test8!

[Gianluca Cecchi]

Hello all,

finally we are!!!

With 2.4.0-test8 we overcome (in my configuration at least) the 2.2.14 in disk performance.

MB: Supermicro P6SBU (Adaptec 7890 on board)
CPU: 1 pentium III 500 MHz
Mem: 128Mb

1x9.1Gb IBM DNES-309170W disk on fast/se channel
4x18.2Gb IBM DNES-318350W on ultra2 channel
The 18.2 Gb disks are in raid0+1 software (md0 is raid0 of two, md1 for the other two and md2 raid 1 of md0 and md1).

This is the output of bonnie in 2.2.14 (excuse me for the format of output):

Version 1.00 ------Sequential Output------ --Sequential Input- --Random-

-Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--

Machine MB K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP /sec %CP

Unknown 1000 7607 96 29971 32 10326 23 8090 98 24738 25 nan -2147483648

------Sequential Create------ --------Random Create--------

-Create-- --Read--- -Delete-- -Create-- --Read--- -Delete--

files /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP

30 169 97 568 99 6000 97 170 97 718 99 337 46

Unknown,1000,7607,96,29971,32,10326,23,8090,98,24738,25, nan,-2147483648,30,169,97,568,99,6000,97,170,97,718,99,337,46

240.88user 557.77system 17:18.92elapsed 76%CPU (0avgtext+0avgdata 0maxresident)k

0inputs+0outputs (791342major+339minor)pagefaults 14swaps



And this is for 2.4.0-test8:

Version 1.00 ------Sequential Output------ --Sequential Input- --Random-

-Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--

Machine MB K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP /sec %CP

Unknown 1000 7900 94 25498 20 11919 17 8579 98 31277 30 nan -2147483648

------Sequential Create------ --------Random Create--------

-Create-- --Read--- -Delete-- -Create-- --Read--- -Delete--

files /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP

30 179 99 +++++ 87 9849 99 181 99 +++++ 102 1270 99

Unknown,1000,7900,94,25498,20,11919,17,8579,98,31277,30, nan,-2147483648,30,179,99,+++++,87,9849,99,181,99,+++++,102,1270,99

227.46user 414.99system 13:35.17elapsed 78%CPU (0avgtext+0avgdata 0maxresident)k

0inputs+0outputs (196major+322minor)pagefaults 0swaps 

A dd of 512Mb takes about 49secs from the raid filesystem to itself!

Long life to Linux! And to you all (don't touch your 8...)!



PS: I will check also behaviour with applications (Oracle 8iR2 and Dev 6i for Linux)...

bye 

Gianluca

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.freebsd.org/pipermail/aic7xxx/attachments/20000913/c0a8630a/attachment.htm