White Stormy
Take that, subspace!
1999TL
New Member
Im pondering if I should use raid or not in my new rig i'm about to build. debate lingers on with myself. Drives are just so big now.
White Stormy
Take that, subspace!
just buy two 1TB+ drives and do raid 1. If a drive DID fail, you have a backup.. and if the backup has a read error and fails.. you're no worse off than if you DIDN'T have a backup solution, except you're out an extra $100-150.
if the backup saves you.. it's probably worth more than $100-150, and that's more likely, so sounds to me like it's still the best answer
Nothing wrong with using RAID 5 on a 3 disk home array or something, with 1TB disks. His point is that we are fast approaching the point that if an entire disk fails, then you will be quite likely to experience a read error during rebuild. Disks are getting bigger, but failure rates are not getting better.
trouphaz
New Member
- Sep 22, 2003
- 2,647
yeah, so what home user is using a 6+1 RAID5 parity group? i wouldn't trust home drives to that little protection. at most i'd go 3+1... though i wouldn't use RAID5 for home anyway. hard drives are cheap and RAID1 (or even RAID1+0) is much better. i can't imagine the rebuild times on many of these low end controllers would be all that great.
This is horseshit. The statistical URE rate might be one bit per 12 terabytes, but that's PER DISK, not for the whole array. In a seven-2TB-disk array with six disks operating simultaneously, you're talking about a URE rate of 1 bit per 72TB, not 12TB.
Anyway, nobody but nobody needs 72TB of data storage, not within reason anyway. Maybe a datacenter has that much capacity, but a datacenter is also going to have far better redundancy than a home user (or even a small office like mine) anyway. We have 4TB of storage and it's half-full -- after five years of continuous operations using data files >1GB apiece, and never deleting anything.
This is nothing more than panic-mongering.
Anyway, nobody but nobody needs 72TB of data storage, not within reason anyway. Maybe a datacenter has that much capacity, but a datacenter is also going to have far better redundancy than a home user (or even a small office like mine) anyway. We have 4TB of storage and it's half-full -- after five years of continuous operations using data files >1GB apiece, and never deleting anything.
This is nothing more than panic-mongering.
This is interesting. What makes you think the rate wouldn't hold across disks? Wouldn't you expect that if you had 1,000,000,000,000 disks, and read 1 byte off each that you would get one bad read?
Limp_Brisket
Active Member
oh, i didn't even notice it was a woman. at least now i can discount everything i read.
edit - it's just a man named Robin, you fooled me!
I say that because while the average number of reads between failures is 12TB, that doesn't mean that the likelihood of failure is identical from Read #1 to Read #n. The MTBF only means that after reading 12TB of data, you have a 50% chance of having suffered one URE; the actual likelihood of failure at any given point in time is described by a hyperbolic curve that starts at 0% at Read #1, crosses the 50% mark at Read #12billion, and trends towards 100% as it approaches Read #Infinity.
Why does that matter? Because if the likelihood of failure per disk starts out infinitesimal and increases as time and wear increase, then that means that 6 disks reading 2TB of data each will not progress as far along the failure-probability curve as a single disk reading 12TB of data would -- the wear is split across multiple disks, so each disk suffers substantially less wear during the rebuild than a single disk would if it had read the same amount of data all by itself. Likewise, the failure-probabilities for all of the drives added together will not equal the failure-probability of a single drive that had read the same amount of data all by itself.
Why does that matter? Because if the likelihood of failure per disk starts out infinitesimal and increases as time and wear increase, then that means that 6 disks reading 2TB of data each will not progress as far along the failure-probability curve as a single disk reading 12TB of data would -- the wear is split across multiple disks, so each disk suffers substantially less wear during the rebuild than a single disk would if it had read the same amount of data all by itself. Likewise, the failure-probabilities for all of the drives added together will not equal the failure-probability of a single drive that had read the same amount of data all by itself.
Last edited:
Limp_Brisket
Active Member
that's his picture in the top right of the page. i thought it was just a stock photo
That article is complete bullshit.
RAID =! backup. You are a moron if you leave all your critical data on one computer without a backup.
RAID =! backup. You are a moron if you leave all your critical data on one computer without a backup.
trouphaz
New Member
- Sep 22, 2003
- 2,647
White Stormy
Take that, subspace!
Limp_Brisket
Active Member
I think we might be getting a little off track.
Good point about Robin Williams though, I forgot about him.
Good point about Robin Williams though, I forgot about him.
trouphaz
New Member
- Sep 22, 2003
- 2,647
actually, i think it has been decided. the bitch who wrote that article is an idiot.