MONZ wrote:
>
> Anyone got experiences with <subj>?
> Docs tells about more uniform performance form outer to inner tracks
> compared to 10K disks, but then again a swap partition on outer tracks
> may perform better on 10K disks, since 15K disks are actually slower on
> outer tracks, according to doc's.
>
> Also, as I understand it, 15K disks do not use uniform lowlevel
> formatting from outer to inner tracks, so a utility like SCU probably
> can't be used on those; guess it can be used on 10K disks to enhance
> performance.
>
> One 15K disk is almost the cost of two 10K disks, so..
Mogens,
Here are some timings for 100 MB sequential transfers from
various block addresses on my 15K rpm Seagate. Looks just
like other disks (degraded performance on outer tracks) only
faster. Througput is varying between 40 MB/sec and 32
MB/sec. The disk is "18 GB" and has 35,843,671 blocks
each of 512 bytes.
$ sg_scan -i
/dev/sg0: scsi0 channel=0 id=0 lun=0 type=0
SEAGATE ST318451LW 0003 [wide=1 sync=1 cmdq=1 sftre=0 pq=0x0]
$ time sg_dd if=/dev/sg0 of=/dev/null bs=512 skip=0 count=200k
real 0m2.542s
$ time sg_dd if=/dev/sg0 of=/dev/null bs=512 skip=20M count=200k
real 0m2.902s
$ time sg_dd if=/dev/sg0 of=/dev/null bs=512 skip=30M count=200k
real 0m3.142s
$ time sg_dd if=/dev/sg0 of=/dev/null bs=512 skip=35M count=200k
real 0m3.434s
Can't imagine why SCU would have any problem with this
disk.
Doug Gilbert
-
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to [EMAIL PROTECTED]
