[mythtv-users] Where are those new 1TB drives?

matthew.garman at gmail.com matthew.garman at gmail.com
Thu Mar 15 13:15:05 UTC 2007


On Thu, Mar 15, 2007 at 11:49:56AM +0000, Stroller wrote:
> The quote "we observe little difference in replacement rates
> between  SCSI, FC and SATA drives" is from a paper which "looked
> at 100,000  drives, including HPC clusters at Los Alamos and the
> Pittsburgh  Supercomputer Center, as well as several unnamed
> internet services  providers". If these are not environments able
> to stress hard-drives,  then I'd be interested to hear of
> environments that do.

Long-range artillery tests?  :)

My take on this is that perhaps, at one point in time, it made sense
for high-end drives (SCSI, etc) to use larger bearings, thicker
platters, etc.  But the mainstream hard drive market has existed now
for what?  At least 20 years.  So I'm sure design and manufacturing
has improved, even on commodity drives.

But I'm sure hard drive manufacturers make a small fortune on their
"enterprise" class drives, so they *need* to be able to
differentiate them somehow.

I think it's akin to selling a passenger car with a frame built for
a tank: *sounds* impressive, certainly costs more, but even under
stressful use, a car's frame should be the last component to fail.

A more precise reason why there's a *visible* difference between
SCSI and SATA/PATA drives is probably the MTBF (mean-time between
failures).  Google or wikipedia MTBF; for hard drives, it is derived
from the MTBFs of all its *components*.  (The exact derivation I
don't recall.)  Point is, bigger/beefier components probably have
bigger MTBF ratings.  But (1) by its own definition, a big MTBF does
not imply that any one drive will last a long time and (2) the study
quoted earlier supports the idea that SCSI's bigger/better
components and MTBF ratings don't mean anything in the "real world",
even under "enterprise class" use.

FWIW, Google also published a similar study (I think it's safe to
assume they have *lots* of hard drives and push them all pretty
hard).

I like this solution to the hard drive reliability problem: use the
least cost drives possible, but always have three live, up-to-date
copies of your data.  (Of course, what you save in cheaper drives
you probably end up paying in electricity, cooling and environmental
impact!)

-Matt



More information about the mythtv-users mailing list