Slashdot Mirror

Information wasn't stored in bubble memory. That was a cooling tank, just like people use liquid cooling today. Mainframe computers used liquid cooling as far back as 1964, almost a decade before the movie came out. Similar to the article, the producers used what they knew about big computers (i.e mainframes) and back then it was liquid cooling.

You can tell the bubbles aren't the memory when near the end of the video it shows the "temperature" getting hotter (the red color) and the bubbles becoming more dense (just like what happens in boiling water).

At least one aspect of an electric car is affected, cooling. As altitude increases, the efficiency of heat-sinks decreases. I hope they added some over-sized cooling fans, Apple didn't.

(sorry for the duplicated posting; the previous one was cut because of problems with the html marks)

In order to obtain a 90% reduction in the energy bill, cooling must account for 90% of the power of the DC. This implies a PUE >= 10. As a reference, 5 years ago virtually any DC had a PUE lower than 3. Nowadays, PUE lower than 1.15 can be obtained easily. As a referecence, Facebook publishes the instantaneous PUE of one of its DC in Prineville, which at the moment is 1.05. This implies that any savings in cooling would reduce the bill, at much, in a factor of 1.05 (1/1.05 = 0.9523).

On the other hand, I believe that this is not the first commertial offer for a liquid-cooled server, Intel was already considering two years ago, and the idea has been discussed in other forums for several years. I can't remember right now which company that was actually selling these solutions, but I believe it was already in the market.

In order to obtain a 90% reduction in the energy bill, cooling must account for 90% of the power of the DC. This implies a PUE >= 10. As a reference, 5 years ago virtually any DC had a PUE instantaneous PUE of one of its DC in Prineville, which at the moment is 1.05. This implies that any savings in cooling would reduce the bill, at much, in a factor of 1.05 (1/1.05 = 0.9523).

On the other hand, I believe that this is not the first commertial offer for a liquid-cooled server, Intel was already considering it two years ago, and the idea has been discussed in other forums for several years. I can't remember right now which company that was actually selling these solutions, but I believe it was already in the market.

There are patents going back a decade pertaining to using microfluidic ducts as a heat transfer mechanism. Every few months now, there's another article on slashdot about one of the chip giants testing out such manufacturing techniques. Just a few links from a quick googling...

http://www.xbitlabs.com/news/coolers/display/20031008155430.html
http://www.electronics-cooling.com/2002/11/electroosmotic-microchannel-cooling-system-for-microprocessors/
http://www.frostytech.com/articleview.cfm?articleid=2424&page=11
http://www.w7forums.com/nanotechnology-delivers-revolutionary-pumpless-water-cooling-t6658.html

Although it may be cost prohibitive, the concepts used when cooling a computer through liquid emersion may do well in this sort of environment. If the expelled heat of the computers is not enough to keep the liquid up to optimal temperature, you can conserve some energy by utilizing the excess heat from the refrigerant system. This method can also be used to raise the "PC-tank" environment up to optimal for a "cold boot" (sorry, could't resist.) The expelled heat of the computers will add to the load of the refrigeration system as a whole and needs to be calculated into the whole power efficiency equation.

Good luck.

Although we call it "water" cooling, there's no reason why you actually have to use water. Other than that it's cheap and easily available. You could alternately fill the system with some sort of nonconductive refrigerant, or even fill it with Freon and keep it pressurized, so that if there was a leak, the fluid would just boil off.

I had an idea a while back to get a big tub of nonconductive liquid and just submerge the whole motherboard in it. I never got around to actually trying it, but it's been in the back of my mind for a while. I think it might affect capacitances and inductances though, and I'm not sure what that would do to (for example) the timing circuits. I got the idea looking back at the old Cray IIs, in which the processor was submerged in some sort of coolant bath, and supposedly the coolant was designed to have a boiling point that would be reached by the hottest parts of the processor -- thereby taking away the heat through the state change. Plus, what I really wanted to immerse the parts in was fluorochemical transformer oils, and let's just say the MSDSs on them will really give you pause, if you ever want to have children, or live that long.

Aside from working with a lot of nasty chemicals, the technology to do direct liquid cooling (where the coolant touches the chips, not where you liquid cool a heatsink that touches the chip) isn't new. Here's a good article on it:
http://www.electronics-cooling.com/Resources/EC_Ar ticles/MAY96/may96_04.htm

I think in a rackmount environment, what you'd want is a dual-loop system. A small loop, contained entirely in the case and permanently sealed, which draws heat away from the chips and takes it to a heat exchanger at the back of the case, which you hook up to the building chilled water supply. That way the only connections (big leakage points) you have are away from the chips, and the only water is back there as well. When you consider the space given over in a modern 1U case to air cooling and airflow, the space savings could be significant, especially if you also offloaded the power supply to the bottom of the rack (water cooled DC power supplies are common laboratory fixtures).

What you're saying is said in lengthier form here.

Thermal conductivity chart.
Please note plane parallel thermsl conductivity of graphite which greatly exceeds diamond.
If you don't want to peruse the linked material...thermal conductivity(W/m - K):copper = 401, diamond = 895, Graphite = 1950.

To answer my own question, a google search turned up this link, which looks pretty informative.

I am quite surprised that on this next "quantum leap" of case design, it wasn't designed around "heat pipes".

There is no reason the entire case itself can't be used as a heat sink, as aluminum is quite thermally conductive. I could only imagine a case that was intentionally designed with a sort of "semi-porous" exterior to facilitate heat transfer and blackbody radiation.

If you want a very technical discussion the article here covers it nicely.

Simply put, full immersion would handle the problem because the fungus would not grow under those conditions. Of course, other factors may make this inconvenient.

OverClockers would likely find the magazine where the article comes from, Electronics Cooling, interesting to read as well.

Check out the Vinny the Vampire comic strip

If you want a very technical discussion the article here covers it nicely.

Simply put, full immersion would handle the problem because the fungus would not grow under those conditions. Of course, other factors may make this inconvenient.

OverClockers would likely find the magazine where the article comes from, Electronics Cooling, interesting to read as well.

Check out the Vinny the Vampire comic strip

If anyone is interested:

I'm curious if there is any application for cooling a CPU below sub-zero for anything other than overclocking (ie- since you are dissapating the extra heat energy, x electrons can get to point y faster). I have no idea what the benefits are but it would be interesting to know.

Link: http://www.electronics-coo ling.com/html/2000_may_a1.html
Of interest: Researchers identified the advantages of operating electronics at low temperatures. These advantages include: faster semiconductor device switching; increased speed due to lower electrical resistance of interconnecting materials; and a reduction in thermally induced failures.
So basically, lower resistance = faster processing speeds, I know I'm stating the obvious here but it's nice to see that there is a more "scientific" approach to this aspect of computing going on.

Running electronics immersed in Fluorinert is an old idea. The Cray 2 was cooled that way. This was more trouble than it was worth, and no later supercomputer used that approach. But Cray built one of the coolest looking computers of all time. The cabinet had windows and the liquid coolant was illuminated. Even the Cray 2 heat exchanger was beautiful.

Here's some actual honest to goodness research on the exact subject, cooling electronics. Imagine that.