Re: Stripe sizing and maxcontig
- From: "John L" <jl@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx>
- Date: Fri, 28 Jul 2006 09:33:25 +0100
"Darren Dunham" <ddunham@xxxxxxxxxxxxxxxx> wrote in message news:j07yg.73397$fb2.28458@xxxxxxxxxxxxxxxxxxxxxxxxxxxxx
Mark <mark.round@xxxxxxxxx> wrote:
First off - the person from the VAR who came round to demo it said that
a RAID 5 volume was perfectly OK, and we wouldn't incurr that much of a
performance hit from choosing that over, say RAID 1+0. Is this correct
? Our databases are much more read-intensive than they are write, but
still... Alarm bells are ringing at this advice even though the array
does have a hefty whack of cache on board.
Could be true. As far as the app is concerned, writes only go to the
cache. Now the destaging to disk will be slower, but the app will never
notice unless the throughput is high enough to cause the cache to fill.
The cache will fill at a lower throughput level with a R5 backing than
with a mirror/stripe backing.
But while the raid controller's battery is charging (which normally takes
around four hours) when the controller will switch to write-through
mode rather than write-back.
This can be configured by the user, of course, but there are data
integrity issues the OP's DBA will need to consider.
--
John.
.
- References:
- Stripe sizing and maxcontig
- From: Mark
- Re: Stripe sizing and maxcontig
- From: Darren Dunham
- Stripe sizing and maxcontig
- Prev by Date: I can see the scsi enclosure but no disks
- Next by Date: ZFS & the automounter
- Previous by thread: Re: Stripe sizing and maxcontig
- Next by thread: Re: Stripe sizing and maxcontig
- Index(es):
Relevant Pages
|
