Oracle FAQ | Your Portal to the Oracle Knowledge Grid |
Home -> Community -> Usenet -> c.d.o.server -> Re: os io unit , db block size relation
FM <fabrizio.magni_at_mycontinent.com> wrote in message news:<3ol2d.333738$5D1.14810469_at_news4.tin.it>...
> oracle_at_pain:/tmp> dd if=/tmp/foo of=/tmp/foo2 bs=1024k count=50
>
> oracle_at_pain:~> iostat -t -x /dev/sda 5
>
> Device: rrqm/s wrqm/s r/s w/s rsec/s wsec/s rkB/s wkB/s
> avgrq-sz avgqu-sz await svctm %util
> /dev/sda 0.00 1.40 0.00 1.00 0.00 19.20 0.00 9.60
> 19.20 50.06 60.00 10000.00 100.00
>
> Time: 08:33:02 PM
> avg-cpu: %user %nice %sys %idle
> 0.50 0.00 4.30 95.20
>
> Device: rrqm/s wrqm/s r/s w/s rsec/s wsec/s rkB/s wkB/s
> avgrq-sz avgqu-sz await svctm %util
> /dev/sda 0.00 2505.80 0.20 67.20 1.60 20584.00 0.80 10292.00
> 305.42 96.54 690.50 148.37 100.00
>...
Using iostat (wsec/s and wrqm/s) or (rsec/s and rrqm/s) values to calculate driver write or read chunk size is more accurate than what I suggested in my last message.
A slightly off-topic question. What if the filesystem block size is not 4k while PAGE_SIZE is 4096? (I had this question before for Solaris http://groups.google.com/groups?selm=b3cb12d6.0203181445.f3d2b36%40posting.google.com)
> In tolga's example he wasn't writing on a real disk so the metrics
> weren't really valid.
You mean he's writing to RAID? Shouldn't RAID be transparent to iostat?
>...
Thanks for the compliments. Some of the stuff on my site is meant to attract recruiters. Real gurus may laugh at it.
Yong Huang Received on Fri Sep 17 2004 - 13:44:08 CDT