> RAID-1 is disk mirror. DPT is cache with RAID. RAID will increase the
> disk performance.
RAID-1 slightly *decreases* write performance, as writes will
complete only after data is written to all disks in the mirror
group. Writes therefore proceed at the speed of the *slowest* disk of
the array. RAID-1 should increase read performance, however.
In any case I would hardly call a poorly conceived and statistically
irrelevant 'benchmark' valid, especially when performed under SCO.
DPT controllers come with a lovely manual which explains in great
detail SCSI, RAID, tagged queueing and all the other basic concepts
that you have demonstrated an inability to comprehend. Please
read it, as these are concepts that every self-proclaimed "System
Administrator" should understand.
> On Mon, 7 Dec 1998, Alan Cox wrote:
>
> > > Open server 5.0.4p, DPT RAID-1, 32 MB cache, PCI, 4.2 GB (A cable)
> > ^^^^^^^^^^^^^^^^^^^^^
You still haven't explained why you are trying to compare SCSI disk
performance under SCO with that of IDE disks on Linux. (Free Clue -
you can't).
Regards,
Damien Miller
| Damien Miller -
| Email: dmiller@ilogic.com.au (home) or damien@ibs.com.au (work)
| WWW: http://www.ilogic.com.au/~dmiller/
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/