Guy wrote:
There's an even *more* interesting paper on
this from a CMU post-doc (math warning...lots
of statistics) that seems to blow away many (if not
all) assumptions about drive failures/quality/raid/etc.
I looked through FAST05 and found a few CMU
papers but nothing that looked like that. If/when
you find the link I'm very interested. My anecdotal
experience is based entirely on a hatred of
unnecessary technology especially RAID controller
failures (there must be a curse: the controller
barfs or loses its config more often than drives
fail!)
Interesting side point: in several case I've
been able to find files (tape images that I made
a decade or two ago) that I had misplaced (not
a drive crash but simply losing them in the shuffle)
by Googling for them and finding a copy of my
tape image, in a directory name that I had made
up twenty years ago, elsewhere on the web!
I was astounded!
Tim.