sata performance with 2.4.32 + latest libata

[Thomas Jarosch]

Hello,

I just tried kernel 2.4.32 + latest libata patches from 2005-12-01
after my performance problems described here:

http://marc.theaimsgroup.com/?l=linux-kernel&m=113171290329839

Now the machine is performing very fast :-)

Results with 2.4.32 + libata latest:
Version  1.03       ------Sequential Output------ --Sequential Input- --Random-
                    -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
Machine        Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP  /sec %CP
proliant         2G 33294  91 57592  20 26778   5 30885  77 56081   5 174.2   0
                    ------Sequential Create------ --------Random Create--------
                    -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete--
              files  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP
                 16 27313 100 +++++ +++ 26421  99 27285 101 +++++ +++ 21477  89
proliant,2G,33294,91,57592,20,26778,5,30885,77,56081,5,174.2,0,16,27313,100,+++++,+++,26421,99,27285,101,+++++,+++,21477,89

Previous results with kernel 2.4.32-rc3:
Version  1.03       ------Sequential Output------ --Sequential Input- --Random-
                    -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
Machine        Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP  /sec %CP
proliant         2G 10523  28 11532   3  7162   1 24271  60 53495   5 164.3   0
                    ------Sequential Create------ --------Random Create--------
                    -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete--
              files  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP
                 16 30498 100 +++++ +++ 27244  98 29630  99 +++++ +++ 11587  46
proliant,2G,10523,28,11532,3,7162,1,24271,60,53495,5,164.3,0,16,30498,100,+++++,+++,27244,98,29630,99,+++++,+++,11587,46

A RAID1 rebuild went from 6-8 mb/s to this:
[=================>...]  recovery = 85.9% (2202816/2562240) finish=0.1min speed=56280K/sec


Thanks Jeff, you rock.

Thomas
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/