Andrew Lentvorski wrote:
Tracy R Reed wrote:
I have heard for years that SCSI drives get the platters and drive heads
that QA'd higher or some such thing. I have always wondered if that was
true. I am skeptical.
It is true that SCSI drives get QA'd better at the high end. I don't
know about SCSI drives at the lower end (ie. 7200 RPM).
They do. Check out the Unrecoverable Bit Error Rate for SCSI vs. IDE
drives. The SCSI drives have an order of magnitude (10x for the
non-mathematically inclined) better (meaning fewer) error rates than IDE
drives. They achieve this through lower areal density and better
electronics for the heads plus more precise tracking.
Besides, it really isn't normally the platter that gives out on a drive.
It is generally the motors or head. Heat causes the motors to burn out
or the semicondtors in the head to fail.
However, a platter failure is an amazing thing to see. 15000 RPM's of
angular momentum turned into linear kinetic energy used to do an
*amazing* amount of damage to the casing. It may no longer be quite so
spectacular now that they use glass platters.
The main reason for stronger QA on SCSI drive is that they push the
thermal envelope *much* harder with 15K RPM drives. Aerodynamic drag is
proportional to v^2. At 15000 RPM, aerodynamic drag (the primary cause
of platter heat) is 4X what it is at 7200RPM. The motor has to be 4X as
strong. And, for SCSI, things are generally qualified for continuous
usage duty cycles.
True. Although some of the newer SATA drives are approaching SCSI
performance and reliability. It seems that Western Digital put a SATA
interface on their SCSI internals, and that Seagate is doing the same.
Gus
--
[email protected]
http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-list
