|
True the math does work out better, but....
It still winds up being effectively three drives, and worse yet from a
write perspective, mirrored. A write to 3 mirrored pairs has to wait
for both of the drives in the mirrored pair to commit the write before
it is released by the storage subsystem to the OS. If you get a large
write that affects more than one of the mirrored pairs, you now have to
wait on all four, or possibly all six drives to complete the write
operation.
Read is some what quicker since it can come from either drive in the
mirrored set.
Since the original question was regarding disk performance a three drive
system (or three mirrored pairs) without much write cache, the system is
going to have problems with any moderate level of I/O.
Jim Oberholtzer
CEO/Chief Technical Architect
Agile Technology Architects, LLC
On 3/29/2011 1:08 PM, Porterfield, Sean wrote:
That math does work out better!
--
Sean Porterfield
-----Original Message-----
From: John Jones
Sent: Monday, March 28, 2011 13:35
To: Midrange Systems Technical Discussion
Subject: Re: Rép. : Question of a Newbie to POWER Systems
Since the OP said
"It contains 3 disks of 140 GB - Which are RAID-1 mirrored totalling 420GB effective."
I was assuming it was 3 pairs/6 physical units.
On Mon, Mar 28, 2011 at 11:49 AM, Porterfield, Sean
JJSince he said RAID1-mirrored, I think it's really 4 drives rather than 3.
If not, that would be the first thing to fix, because a disk is
unprotected.
--
Sean Porterfield
--
4 Out of 3 people have trouble with fractions.
As an Amazon Associate we earn from qualifying purchases.
This mailing list archive is Copyright 1997-2024 by midrange.com and David Gibbs as a compilation work. Use of the archive is restricted to research of a business or technical nature. Any other uses are prohibited. Full details are available on our policy page. If you have questions about this, please contact [javascript protected email address].
Operating expenses for this site are earned using the Amazon Associate program and Google Adsense.