View Single Post
  #117   Report Post  
posted to rec.boats
Wayne.B Wayne.B is offline
external usenet poster
 
First recorded activity by BoatBanter: Jul 2006
Posts: 10,492
Default A nice apple story

On Thu, 17 Nov 2011 12:10:52 -0500, wrote:

On Thu, 17 Nov 2011 08:00:12 -0500, Wayne.B
wrote:

On Thu, 17 Nov 2011 02:18:02 -0500,
wrote:

The 33xx drives just failed hard but they really didn't fail that
often. Most failures didn't actually lose data unless you had a head
crash.


===

Once your DASD farm got up to 1,000 plus drives of slightly aging
33xx's, unit failures became almost a weekly occurrence, each one
requiring a huge amount of time and effort to recover the data.


Nashua 3330 disk packs ;-)

I didn't really see that many HDA related failures and we had lots of
DASD.
I agree the RAID 5 RAMAC made data loss a lot more rare but that
really sets people up for a disaster. The more frequently you have a
failure, the better you keep your stuff backed up.


====

The 3350s were better but not immune either. I think IBM was quoting
a MTTF of something like 7 years but that is a statistical mean.
Some fail quicker. Even at an average of 7 years, once you get to
several thousand units the probability of experiencing a failure in
any given week is quite high.