PCBUILD Archives

Personal Computer Hardware discussion List

PCBUILD@LISTSERV.ICORS.ORG

Options: Use Forum View

Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Tom Turak <[log in to unmask]>
Reply To:
PCBUILD - Personal Computer Hardware discussion List <[log in to unmask]>
Date:
Mon, 11 Jun 2001 12:19:49 -0400
Content-Type:
text/plain
Parts/Attachments:
text/plain (50 lines)
MTBF does not refer to the expected life of each individual drive.  It is
based on the experience of an installed population of like drives.  It is
used to predict the failure rate for a model of drives within their useful
lifespan.  I used to think it meant if 200,000 drives operated for one hour,
one would fail, but now I'm not so sure.  As for how long the average drive
will last, the answer is it will last almost 100% of its useful life, since
MTBF does not count failures beyond the useful lifespan, and failures still
occur, so it follows not all drives will reach the end of their useful life,
but most do.  Now, a lot I have read on the subject goes right over my head,
but it seems a higher MTBF equates to a larger percentage of units exceeding
their useful life, which means your chances of getting a lemon go down.  I'm
too lazy to get a confirmation, but I think I remember ultra-scsi drives had
a useful life of 5 years, the newer ide models out now could be longer.
Here is a web article
http://alumni.caltech.edu/~rdv/comp-arch-storage/FAQ-2.12.html

I'm going to assume both 5400 rpm and 7200 rpm have the same useful
lifespan.  Then an equal MTBF rating would mean they are equally reliable.
Tom Turak

-----Original Message-----
From: Carroll Grigsby [mailto:[log in to unmask]]
Sent: Saturday, June 09, 2001 11:56 AM

Joel:

A minor correction: MTBF is the Mean Time Before Failure, not Maximum.
It is a calculated parameter based on extensive testing of the various
components that are used in the drive, and is an estimate of how long an
"average" hard drive of this design would last under typical operating
conditions. In theory, if we tested a large number of these drives to
failure, the average of the failure times would be 150,000 hours -- some
will fail sooner, and some will last longer. In practice, of course, no
hard drive is going to survive 17+ years.

Hard drive manufacturers sometimes list other reliability parameters
that may be of more relevance. I have a Maxtor drive that is rated as
being capable of at least 50,000 stop/start cycles. It also has a rated
MTBF of 500,000 hours -- that's about 57 years! If we believe both of
these numbers, then the 500,000 hours life will occur only if this drive
experiences one stop/start cycle every 10 hours. Oh, by the way, the
warranty is for three years. Fujitsu quotes similar numbers (500,000
MTBF, 40,000 stop/start cycles), but they also state that the component
life is 5 years. I expect that the actual limiting factor is read/write
cycles -- but neither of these manufacturers provides this rating.

                         PCBUILD's List Owners:
                      Bob Wright<[log in to unmask]>
                       Drew Dunn<[log in to unmask]>

ATOM RSS1 RSS2