[plug] Using a HDD with Badsectors

[Bernd Felsche]

"Mark J Gaynor" <mark at mjg.id.au> writes:
>On 26/11/2006 at 12:42 PM Chris Caston wrote:

>>IMHO with the falling cost of storage these days I would just replace
>>the drive.

>I agree with Chris, it is not worth the trouble these days. If you
>can keep the ambient temp of your environment round 18 Celsius they
>will run for years and attain the manufacturers MTBF figures. (Mean
>Time Between Failures) IFIRC at plus two degrees you have lost
>about 20-25% of the expected life of the device, and every degree
...

Only HALF the drives will attain the MTBF. The other have will, by
definition, have failed. It's the centre of the "bell curve" that is
characteristic of the failure rate.

Server drives have MTBF of the order of 5 million hours. Well over a
500 years. It's not an exercise in one-upmanship that has the
manufacturers build such drives; it's trying to push the centre of
the bell curve into the distant future so that the rate of failure
in operational environments is 

>Ever wandered why machine rooms are so cold! . . . .  To get the

I've not wandered. I usually know eher I'm going in a machine room.

Machine rooms are traditionally cold because the old iron produced a
lot of heat from a few functional elements.

There is SFA difference between MTBF @ 16 degrees C; the coldest
you'd be allowed to run a room under ECE regulations; and 25 degrees
C; the highest "comfort" temperature for humans, given adequate
airflow. Yet the airconditioning cost of 16 vs 25 degrees C in an
average Australian office building during summer is "astronomical".

The problem relates to the need to remove heat that's inevitably
produced by the equipment when operating.

Unless you're buying real, server-class hardware, it's rare to find
equipment that's designed and built for optimum heat dissipation.
And it's a marvel to behold consumer installations of heat-sensitive
equipment.

IME; the worst enemy of electronics is an interior designer.

>best performance out of the equipment, and it is not restricted to
>computers either, anything electronic.

It's not just the mechanical and magnetic bits that wear out.
Solid-state electronics from the transistor to multi-mega-junction
chips will eventually "wear out". It's metalurgical.
-- 
/"\ Bernd Felsche - Innovative Reckoning, Perth, Western Australia
\ /  ASCII ribbon campaign | "If we let things terrify us,
 X   against HTML mail     |  life will not be worth living."
/ \  and postings          | Lucius Annaeus Seneca, c. 4BC - 65AD.