[Beowulf] 512 nodes Myrinet cluster Challanges

[Jim Lux]

At 01:17 PM 5/2/2006, Bill Broadley wrote:

> > blower, rather than a bunch of 40mm axial/muffin fans.  a much larger 
> cluster
> > I'm working on now (768 nodes) has 14 40mm muffin fans in each node!  while
> > I know I can rely on the vendor (HP) to replace failures promptly and 
> without
> > complaint, there's an interesting side-effect: power dissipation.  of 
> 12 fans
> > pointing at the CPUs are actually paired inline, and each pair is rated to
> > dissipate up to 20W.  so a node that idles at 210W and 265W under full load
>
>Wow, 10 watts per 40mm fan?  Really?
>That would be a square of fans, er 16cm on a side
>
> > can easily consume 340W if the fans are ramped up.  ouch!
>
>75 watts difference, hrm, 75/12 = 6.25 watts difference per fan between
>the low rpm cool state and the hot state...  larger than I'd guess.

Don't forget to factor in power supply and *fan controller* efficiency.  A 
fan drawing, say, 6 Watts DC might wind up drawing 8 or 9 W from the AC 
line.  The basic PC power supply is probably only 80-85% efficient 
(particularly at lower loads), and the fan speed controller is probably 
some inexpensive PWM modulator, itself not particularly efficient, esp at 
low loads.


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
tel: (818)354-2075
fax: (818)393-6875