Hacker News Reader

Comment by ajcp | Hacker News Reader

Semi-related: I find the seemingly simple (or basic?) issue of cooling to be really interesting. Mainly because so much energy, effort, and cost is spent on cooling, and our tech and manufacturing industries depend so much on it, yet I don't see anyone trying to "disrupt" it. I very well could not be plugged in enough to know if anyone is or isn't though.

That being said it seems like one could offer cooling solutions that would cost them very little after initial setup, like providing infrastructure in places that are "naturally" very cool. This would probably be either deep in the oceans (I believe Microsoft piloted deep-submerged servers) or high up in the atmosphere, or even space itself.

Just a thought exercise I suppose...

View on HN

I fail to see how you could "disrupt" something like cooling solutions in the way I believe your post implies.

First, the ability to remove large volumes of heat from a DC (or analogous plant) is largely dependent on the local geography/environment. In modest cases, that may come down to average ambient temperatures and the cost of electricity. In more extreme cases, access to a massive heat sink such as a body of water may be necessary.

Second, the technology used to evacuate heat is very mature. The modern world depends on HVAC and has for a very long time. While there are incremental advances such as new refrigerants and compressor technology, there is always a cost to performance tradeoff.

Note that the theoretical limit on cooling performance (CoP) of a Carnot engine is ~7.8. Check out this report by the IEA (page 43) https://iea.blob.core.windows.net/assets/0bb45525-277f-4c9c-...

If you are trying to cool something to <10K, I don't think it matters all that much what temperature you start at. Unless you start at space, which is already that cold (in shadow) but sending and operating stuff in space is hardly a cheap alternative.

Yeah that's was what I suspected. Outside of space-based infrastructure or at massive crush-depths in the ocean the "naturally" cool places wouldn't really make a dent in the required temperatures.

I don't wonder then if, when the cost of build goes down and speed of transmission goes up, space based infra for these things will be a big industry.

If anything it’s a lot harder to cool things in space because you are limited to radiating it.

The ISS has massive radiators almost as large as the station itself at 1800 square ft of radiator panels.

That makes sense. I don't know why I just assumed the vacuum of space would provide a perfect and passive cooling environment, but of course that's not the case.

There are a lot of examples in movies and the zeitgeist, even where other harder Sci fi Is obeyed, so it's hardly surprising if you'd never stopped to think about it. Welcome to the 10,000.

Qua intuition: Think vacuum flask aka thermos flask. (though these also try to reduce radiative cooling with a mirror-finish)

Convective cooling is a very efficient form of cooling, but obviously doesn't exist in a vacuum. Typically people resort to radiative cooling in space. This requires larger radiators.

Cooling in space is actually very difficult. While unlit space can be thought of as "cold" (ie you're in thermal equilibrium with a very cold resevoir) vacuum is an incredible insulator so once you get any heat coming in (including the onslaught of heat from the local giant nuclear fireball) it's very difficult to dissipate.

More generally, ambient temperature is irrelevant, what matters is heat transfer rates. If you just stick something someplace cool, it'll heat up. You need something like a flowing river to carry the heat away. This is one reason that things like power plants tend to be built on rivers. For things like computing, total heat production is much less of an issue and can easily be conveyed away from the building by airflow. The difficulty is cooling within the building, as tightly packed server racks are not very conducive to airflow, so you need active cooling.

Disrupting the 2nd law of thermodynamics has proven hard, yes. I for one wouldn’t bet against it.

Reply on news.ycombinator.com