spreading the heat?

spreading the heat?

[Folkert van Heusden]

I was wondering: if a quad core runs 2 processes at max cpu usage, won't
that wear out the chip? As the heat is not egally spread out, some parts
will expand more then other parts, may give damage won't it?
So what I was suggesting maybe if that is the situation for longer time,
maybe such processes should be scheduled to other cores as well?


Folkert van Heusden

-- 

Multitail - gibkaja utilita po sledovaniju log-fajlov i vyvoda
kommand. Fil'trovanie, raskrašivanie, slijanie, vizual'noe sravnenie,
i t.d.  http://www.vanheusden.com/multitail/
----------------------------------------------------------------------
Phone: +31-6-41278122, PGP-key: 1F28D8AE, www.vanheusden.com

Re: spreading the heat?

[Lennart Sorensen]

On Wed, Nov 21, 2007 at 09:37:36PM +0100, Folkert van Heusden wrote:
> I was wondering: if a quad core runs 2 processes at max cpu usage, won't
> that wear out the chip? As the heat is not egally spread out, some parts
> will expand more then other parts, may give damage won't it?
> So what I was suggesting maybe if that is the situation for longer time,
> maybe such processes should be scheduled to other cores as well?

If intel didn't design it to survive having a non SMP OS booted on it
using just one core 100% (say you boot DOS which doesn't have any
idle/sleep stuff in it, just actively spiing looking for input), then
the design is wrong.

I highly doubt that so I wouldn't worry about it.  After all that metal
bit on top of the cpu is called a heat spreader for a reason.  I would
be surprised if you can get more than 10 degrees C difference between
different parts of the core, although that's just a guess.

--
Len Sorensen