[Beowulf] IBM Sequoia

Nifty Tom Mitchell niftyompi at niftyegg.com
Thu Feb 5 12:57:38 PST 2009


On Wed, Feb 04, 2009 at 02:47:12PM -0800, David N. Lombard wrote:
> On Wed, Feb 04, 2009 at 12:58:27PM -0700, Robert G. Brown wrote:
> > On Wed, 4 Feb 2009, John Hearns wrote:
> > > 2009/2/4 Bruno Coutinho <coutinho at dcc.ufmg.br>:
> > >>>
> > >> Each compute card could have a octa core processor and some memory chips.
....
> > >> But cool this thing will now be easy.
....
> > > Seymour Cray is laughing out loud on his personal cloud of freon
> > > vapour right now.
> 
> Fluoroinert perhaps.
> 
> > Immersion in liquid nitrogen.
> 
> Some of the dinosaurs^Wmore experienced list members will remember the
> ETA-10 which used LN2 in all but their air-cooled "Piper" system.  My
> undertanding at the time was the Piper was a vehicle to consume the 
> Honeywell VHSIC chips not suitable for the LN2-cooled versions.
> 
> For today's more adventerous, less exotic mineral oil is being used, e.g.,
> as shown in <http://www.pugetsystems.com/submerged.php>

The shock induced by a liquid nitrogen bath might risk damaging packages,
PWB, traces, sockets and more.  All of which would mandate managed
cool down and warm up schedules.   Also I do not know what the transfer
function of liquid nitrogen is.  Is the nitrogen boiling or being pumped
(dual phase or single phase), pressurized....  Since pure nitrogen can
suffocate, a room full of it might incur OSHA issues well beyond a dewer
of LN in a lab.  Also most components are not designed or characterized
for operation at −321 °F.

How large was the compressor and cooling tower facility for the ETH-10?
One web page commented that it was much larger than the machine room....

Does anyone recall the temp that Cray kept their Fluorinert at
during operation. And then was chilled water/ brine used to cool the 
Fluorinert.

Fluorinert has an interesting property in that it boils at about 125°F
and the pressure in the container can then be a strong indicator of
system overheating.  Also, boiling into a sufficient buffer volume can sync a
lot of heat in the case of power/ pump failure.

In my youth I worked at an ice house with a massive ammonia based cooling
system.   Given the green movement I suspect that some on this list will
see central brine and ammonia systems as primary chiller resources in
their future.

This is an interesting topic to discuss but when distilled to its
core it mostly boils down to lots of hot air.


-- 
	T o m  M i t c h e l l 
	Found me a new hat, now what?





More information about the Beowulf mailing list