Re: Disk performance
Re: Disk performance
- Subject: Re: Disk performance
- From: Sam Vaughan <email@hidden>
- Date: Sat, 3 Nov 2007 09:20:16 +1100
On 02/11/2007, at 7:21 PM, Anton Altaparmakov wrote:
Yes you did. You want "skip=" not "seek=". Using seek will seek
inside /dev/null not inside /dev/rdisk0s2 thus all your reads are
from the same place. Try it again with skip instead of seek and
you should find the same results as the other test with individual
partitions.
Thanks, Anton. Sure enough it's staring me in the face in the man page!
seek=n Seek n blocks from the beginning of the *output* before
copying.
Skip and iseek are in units of the block size specified, in this case
1m, so by changing it from 'g' to 'k' I get what I was looking for:
$ for i in `jot 10 0 900`; do
> echo -n offset "${i}GB: "
> sudo dd if=/dev/rdisk0s2 of=/dev/null bs=1m count=1k skip=${i}
k 2>&1 | \
> perl -ne 'if (/\((\d+)/) {printf "%.1fMB/s\n", $1 / (1 << 20);}'
> done
offset 0GB: 82.4MB/s
offset 100GB: 84.3MB/s
offset 200GB: 81.2MB/s
offset 300GB: 78.9MB/s
offset 400GB: 74.6MB/s
offset 500GB: 71.0MB/s
offset 600GB: 66.7MB/s
offset 700GB: 61.6MB/s
offset 800GB: 54.6MB/s
offset 900GB: 46.7MB/s
$
Nice results.
Given that the innermost GB I can read at offset 930GB comes in at
about 43MB/s, and circumference is proportional to radius, I guess
these results mean that the innermost track is about 3.5/2 * 43/83 =
0.9 inches out from the centre of the disk; i.e. just over half-way.
Seems a fair way out...
Then again I don't get 10/7.2 * 83 = 115MB/s from the first track of
the 10k Raptor, so there's obviously more to it than my gross
simplification.
Thanks again Anton,
Sam
Attachment:
smime.p7s
Description: S/MIME cryptographic signature
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Filesystem-dev mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden