Re: Raid ?
- From: Jaimie Vandenbergh <jaimie@xxxxxxxxxxxxxxxxxxx>
- Date: Fri, 06 Jan 2006 20:19:53 GMT
On Fri, 6 Jan 2006 14:34:21 +0000 (UTC), "Rob Nicholson"
<rob.nicholson@xxxxxxxxxxxxxxxxxxxxxxxxxx> wrote:
>> Yeah but RAID-0 with 2 disks is twice as likely to fail :-)
>
>Hmm, not sure I agree with your math :-) Let's say you've got two 120MB
>RAID-0 drivers giving you 240MB total. Sure, one half fails and you're
>screwed. So if you weren't going to use RAID for speed, then you'd probably
>get one bigger hard disk. Therefore, you've still got the same chance of it
>failing.
If each drive has a 1-in-100,000 chance of failing every hour, then
every hour one (250gig) drive has a 1-in-100,000 chance of dying.
Two (120gig) drives in a raid0 array, each at 1-in-100,000 chance per
hour --> 2 in 100,000 chance of _one_ of them dying per hour.
If you push it out further to make it more obvious, a raid of 100,000
drives would have one die per hour, on average.
>As just to tempt fate, my IDE RAID array did fail yesterday
Bad luck. I've had to RMA a couple of 250gb WD SATAs (the JD models)
at work in the last year, never any other WD drives. Hmm.
>all I've lost is a few days of stuff, mainly email. And because lovely GMAIL
>keeps it all your old email, it's was easy to get that back.
Splendid!
Cheers - Jaimie
--
"Every Little Thing She Does Is Sufficiently Advanced Technology"
.
- Follow-Ups:
- Re: Raid ?
- From: Rob Nicholson
- Re: Raid ?
- References:
- Raid ?
- From: Redman
- Re: Raid ?
- From: Jaimie Vandenbergh
- Re: Raid ?
- From: Redman
- Re: Raid ?
- From: Jaimie Vandenbergh
- Re: Raid ?
- From: Trevor Best
- Re: Raid ?
- From: Rob Nicholson
- Re: Raid ?
- From: Trevor Best
- Re: Raid ?
- From: Rob Nicholson
- Raid ?
- Prev by Date: Re: PCI IDE controller card for a 1996 PC... recommendations? TIA
- Next by Date: Re: o/t USB1.1 or 2.0
- Previous by thread: Re: Raid ?
- Next by thread: Re: Raid ?
- Index(es):
Relevant Pages
|