HD failure rates?

Y

Yamaki

Gawd
Joined
Dec 27, 2002
Messages
741
I've recently been thinking of purchasing two of those WD 640GB drives for a speedy RAID 0 array, but I'm sort of in a dilemma. RAID 0 is great, but I don't want to lose all of my data if a disk fails. The thing is, it seems to me that hard disks nowadays rarely fail (I've only had one fail on me ever, and that was back in 1999). Also, my data is just typical stuff...music, movies, games. Nothing that would cause a huge problem if it was lost--it would just suck.

Would you guys say that HDs nowadays are pretty much guaranteed not to fail within 2-3 years? If so, RAID 0 is a perfect option for me. Thanks!

-Yama
 
No. You have been lucky that you have not had a failure. There is no guarantee of non-failure. Speedy R0 arrays are fine, just don't keep any non-backed up data on one.
 
I've had 2 fail on me in the last 2 years (one was DOA and the other died in a couple of months), and I've been building for the last 7 years. I do have another one I cannot secure erase, so I'm pretty sure that one has bad sectors and needs to be RMA'd.
 
The only guarantee with a hard drive is that it will fail eventually, and by eventually I mean sometime between the moment after it's produced and decades down the road. I've had drives dead out of the box, and on the other hand I still have drives over 20 years old that work. If you don't want to lose data, you need to backup regularly. The Google hard drive study posted above provides a good example, drives can fail at any time.
 
you should always backup important stuff, period
for what its worth, ive only had 2 hard drives go bad over 15+ years of PC ownership. Figure about 20-30 personal drives owned. I sold off alot of the old ones, so I dont know how long they lasted, but im still running some old 10 and 20GB drives in some systems every day and they still work great.

that being said, always BACKUP your stuff.
 
Home PC failure rates are much different than Google failure rates. Home PCs are either not on all the time or have the hard drives spun down most of the time. Google drives spin full time.

I have never had a hard drive problem that was caused by the hard drive. (20 years of computer usage in our business - 8TB of current storage.)

I have a controller that supports staggered spin up and it tends to kill WD hard drives if you set the controller for staggered spin up.
 
GeorgeHR said:
Home PC failure rates are much different than Google failure rates. Home PCs are either not on all the time or have the hard drives spun down most of the time. Google drives spin full time.

I have never had a hard drive problem that was caused by the hard drive. (20 years of computer usage in our business - 8TB of current storage.)

I have a controller that supports staggered spin up and it tends to kill WD hard drives if you set the controller for staggered spin up.
Click to expand...

Well I suppose that depends on your definition of "home." Many of us here consider themselves "prosumers" that neologism for professional-consumer. I and many others, have fileservers, computers, NAS's etc that remain on 24/7/365. My laptop has been on nearly 24/7/365 except when it is in transit, and its going on 2 this fall, with little to no failures, even tho laptops are much more prone to errors. Again, YMMV.
 
A lot of hard drive failures are probably due to rough shipping and handling. That's what http://faq.storagereview.com/tiki-index.php?page=BrandMostReliable says and they are probably correct (although some Maxtors were getting horrid reviews for awhile). WD's new drives with SecurePark should help cut down on the bumping around defects. I always liked retail drives from Newegg because they came packed better than the OEM drives. I think most failures are in first months too so if you had a couple drives for awhile that seem to be working thats a plus.
 
I've had several RAID failures since I first started using it in about 1998.

1. RAID 0, IBM Deathstar 45GXP, drive failure
2. RAID 0, IBM Deathstar 75GXP, drive failure
3. RAID 1, IBM Deathstar 45GXP (replacement), drive failure
4. RAID 1, IBM Deathstar 75GXP (replacement), drive failure
5. RAID 0+1, IBM Deathstar 45GXP + 75GXP (replacements), drive failure
6. RAID 1, Asus P4C800 mobo, RAID controller hiccup for both drives - used GetDataBack to recover data.
7. RAID 5 array on MSI 6150 chipset mobo, worked fine for a year then became flaky, dropping a drive once a month. Abandoned the array.

1-5 were two 45GXP's and two 75GXP's that I eventually replaced each drive twice under warranty before throwing all the drives away after the RAID 0+1 array couldn't even keep up with the drive failures. I learned much about data backups during that era.
 
^^ Haha.. that looks almost exactly like my failure rate.. IBM Deathstars.. I have like 2 left that are still alive..Other than that I've only had a Maxtor drive go..
 
I have had raid0 failures without the disks themselves being bad. Integrated and software raid controllers suck imo. As already said, a good backup strategy certainly reduces the potential losses of a raid0 setup.
 
I've lost 2 IBM Deskstars in the past several years, but I've lost 4 Maxtor DiamondMax drives too. After losing my backup with 600 GB of data, programs, music, etc., I've lerned to back my system up continuously. Losing data sucks.
 
You must log in or register to reply here.
Back
Top