Stress Testing Hard Drives

[Oren Beck]

In regard to the concept of drive degradation being "hidden" by controllers.
It seems to be a needful evil for the current applied art and science of
rotating magnetic media at these price points. Yes, I suspect we "could"
make media, heads, positioning methods et all that would be "true" non
contact under most practical conditions.

The game is a paraphrasing of an older tech rubric- Cheap- Fast- Stable=pick
any 2.

So the SMART ancestral  concept  was actually devised as an end run around
those rules.
Setting a scoring metric for considering a "data element" as pass/fail is
still a moving target.
Set your metrics too conservative and a drive becomes so apparently crappy
that you would consider it as unsafe to trust. Set the metrics too
optimistic and you replicate the ZIP drive impact- cheap- comparably fast
for a floppy replacement- but dangerously unstable at times.
Some comments here invoked the Gibson drive tools. There is a Zip issue
called Click of Death.
Mr Gibson did some excellent research into the cause, prevention, and
recovery from the issue
That body of work gives some insight to how modern drives work and fail.

There's an element of Schrodinger's Cat indeterminate factors inherent in
dimensions approaching or even exploiting quantum effects. Modern heads owe
their ancestry of late more to a hall effect transistor than a magnetic
coil. That allows flying head gaps to be larger for the same signal than
older heads. Which paradoxically makes a head crash more devastating due to
higher impact forces if from nothing else than SPEED! I could go on at
length but It would be out of scope for the concept this thread is working
out. Here's my topical constructive query .

In the form of a friendly challenge- Let's see who can share with our group
the most useful Open Source usable from Linux tools to interrogate drive
controllers as to  true drive metrics.

Show us the code....

Oren
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://kclug.org/pipermail/kclug/attachments/20070706/40704047/attachment.htm