sslug-teknik team mailing list archive

[Gert Holtoft <gert@xxxxxxxxxx>]

Onsdag 02 maj 2007 11:56 skrev E. Sjørlund:

> hdparm -t /dev/dindisk
> vi kunne sige om ydelsen er god. Er den det, saa bliv der.

Det her ser ikke så ringe ud, synes jeg:
/dev/md1:
 Timing buffered disk reads:  288 MB in  3.01 seconds =  95.79 MB/sec
dulle:~ # hdparm -t /dev/md0

/dev/md0:
 Timing buffered disk reads:   86 MB in  1.41 seconds =  60.80 MB/sec
dulle:~ # hdparm -t /dev/md2

/dev/md2:
 Timing buffered disk reads:  330 MB in  3.01 seconds = 109.77 MB/sec
dulle:~ # hdparm -t /dev/md3

/dev/md3:
 Timing buffered disk reads:  284 MB in  3.02 seconds =  94.04 MB/sec
dulle:~ # hdparm -t /dev/md4

/dev/md4:
 Timing buffered disk reads:  282 MB in  3.00 seconds =  93.95 MB/sec
dulle:~ # hdparm -t /dev/md4    

Sammenlignet med diskene hver for sig:

/dev/sda:
 Timing buffered disk reads:  182 MB in  3.00 seconds =  60.58 MB/sec
dulle:~ # hdparm -t /dev/sdb

/dev/sdb:
 Timing buffered disk reads:  186 MB in  3.01 seconds =  61.87 MB/sec
dulle:~ #

mvh/gert


-- 
While anyone can admit to themselves they were wrong, the true test is
admission to someone else.