[5411] in linux-scsi channel archive
Re: Dumb question: Which is "better" SCSI or IDE disks?
daemon@ATHENA.MIT.EDU (Mike Jagdis)
Mon Dec 14 11:50:45 1998
Date: Mon, 14 Dec 1998 14:53:18 +0000 (GMT/BST)
From: Mike Jagdis <mike@roan.co.uk>
To: Eric Lee Green <eric@linux-hw.com>
cc: Gregory Maxwell <linker@z.ml.org>, Christopher Smith <cbsmith@envise.com>,
Leonard Zhang System Administrator ISD RVIB <leonardz@rvib2.rvib.org.au>,
linux-kernel@vger.rutgers.edu, linux-scsi@vger.rutgers.edu
In-Reply-To: <Pine.LNX.4.04.9812121654490.5572-100000@england.local.net>
On Sat, 12 Dec 1998, Eric Lee Green wrote:
> On Wed, 9 Dec 1998, Gregory Maxwell wrote:
> > What? Join the real world but. Raid-1 doubles read throughput and the
> > slowdown of writes is negligable.
>
> Sorry, but I don't find that in actual benchmarks. Perhaps it is true
> with some RAID controllers, but definitely not with the ICP-Vortex GDT
> 6517. I get the same throughput from RAID1 on that setup that I get
> from the bare Seagate Cheetah on that setup (no RAID).
It's perhaps worth mentioning that different hard disks can
behave *very* differently. I have two controllers with a
variety of matched pairs of RAID0 striped hard disks. All the
disks give 6-7MB/s individually using hdparm. The IBM pair
happily give ~11MB/s regardless of the RAID chunk size. The
pair of Seagates *only* get the expected ~11MB/s when using
a chunk size of 32k. Anything more or less and the performance
collapses back to around ~7MB/s. It looks like (these) Seagates
have "strange" caching strategies :-).
Moral: RAID is not a panacea. As with all IT you can't take
anything on spec. Every "fact" has to be checked and double
checked...
Mike
--
A train stops at a train station, a bus stops at a bus station.
On my desk I have a work station...
.----------------------------------------------------------------------.
| Mike Jagdis | Internet: mailto:mike@roan.co.uk |
| Roan Technology Ltd. | |
| 54A Peach Street, Wokingham | Telephone: +44 118 989 0403 |
| RG40 1XG, ENGLAND | Fax: +44 118 989 1195 |
`----------------------------------------------------------------------'
-
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@vger.rutgers.edu
