[X4U] Re: Lifespan of Harddrives? (a bit off-topic)

[Bill Bauldry]

On 2/15/06 11:43 AM, "x4u-request at listserver.themacintoshguy.com"
<x4u-request at listserver.themacintoshguy.com> wrote:

>> From: Bill Bauldry <BauldryWC at appstate.edu>
>> I¼ve been amused for a long time by manufacturer¼s claims for a hard drive¼s
>> `Mean Time Before Failure` (MTBF) stats used in product data sheets. A top
>> high-end drive company lists 1,400,000 hours MTBF on a new SATA drive. It
>> doesn¼t take rocket science to analyze this claim:
>>     1,400,000 hours = 58,333.3 days = 159.8 years!
>> I¼ve never met a hard drive yet that would outlast me by that much...
>> 
>> Regards,
>> Bill
> 
> You've never thrown away a piece of kit that still had a working hard drive?
> 
> David
> -- 
> David Ledger - Freelance Unix Sysadmin in the UK.
> Chair of HPUX SysAdmin SIG of hpUG technical user group (www.hpug.org.uk)
> david.ledger at ivdcs.co.uk
> www.ivdcs.co.uk


I've definitely dumped some good hard drives - in fact I've got a perfectly
good PowerMac 8500 sitting unplugged under my desk waiting for me to figure
out what to do with it.

But - I know how the MTBF figures are calculated - and they are wrong. It's
an improper application of "statistical independence" that gives the silly
figures. Think of it this way: if the mean life of a computer is 5 years and
the MTBF of a hard drive is 160 years, it should be such an incredibly rare
event for a hard drive to fail that Norton, Alsoft, Prosoft, Micromat, etc.,
would never have gone into the business. Manufacturer's claims
notwithstanding, a Google search gives over 2.3 million hits for "hard drive
repair."

Regards,
Bill
______________________________________
Wm C Bauldry, PhD
Professor and Chairperson
Department of Mathematical Sciences
Appalachian State University
Boone, NC 28608-2092
_____________________
mailto:BauldryWC at appstate.edu
http://www.mathsci.appstate.edu/~wmcb/
______________________________________