[Beowulf] /. Cooler room or cooler servers?
James.P.Lux at jpl.nasa.gov
Fri Apr 8 09:35:52 PDT 2005
At 07:47 AM 4/8/2005, Joshua Halpern wrote:
>Jim Lux wrote:
>>At 04:27 PM 4/7/2005, Mark Hahn wrote:
>>This what I would imagine off hand... Comparable computational power of
>>a given IC fabrication technology (feature size/voltage/etc) will
>>dissipate similar amounts of power. And this is what you said you measured.
>Shifting away from silicon to large bandgap semiconductors could be a way
>out. In principle such devices could operate at much higher
>temperatures. The development push is coming from operation in hostile
>environments, such as car engines and stand alone operations where cooling
>is not available.
And in space.. Higher operating temperatures are very good, because you
have to get rid of heat by rejecting it with a radiator to cold
space. Moving the temperature of the radiator up from 300K to 400K helps a
whole bunch (that T^4).. you can make the radiator 1/3 the size.
>At this point I would guess that the best material would be silicon
>carbide rather than gallium nitride. This is all pie in the sky right
>now but there are no first principle obstacles just money and time.
Well... there are a few yield problems, and so far, many hundreds of
millions (if not billions) of dollars have been spent on trying to improve
it without huge success. Don't expect to be seeing 0.25 micron linewidth
12" wafers of SiC coming off the line anytime soon. The big use of SiC now
is White LEDs, I think.
James Lux, P.E.
Spacecraft Radio Frequency Subsystems Group
Flight Communications Systems Section
Jet Propulsion Laboratory, Mail Stop 161-213
4800 Oak Grove Drive
Pasadena CA 91109
More information about the Beowulf