News for hackers. Stuff that matters.
Re:Microsoft's underwater data centre resurfaces after two y... by StillBored on Monday September 14, 2020 @04:55PM (#24475051)
In none of the articles do I see the power consumption of the pod..

But it looks packed with a lot of servers, and the surface area is quite minimal given there don't appear to be any cooling fins/etc.

So, that many servers are definitely many KW of power, and its all being conducted away via what appears to be a fairly minimal surface area. So the problem probably isn't the exterior so much as the interior which appears to mostly be a air->paint interface.

So, whats the cooling mechanism here, or are the servers that low power?

Re:Microsoft's underwater data centre resurfaces after two y... by petascale on Monday September 14, 2020 @05:36PM (#24475445)
240 kW according to Microsoft: https://natick.research.microsoft.com/

> [They] adapted a heat-exchange process commonly used for cooling submarines to the underwater datacenter. The system pipes seawater directly through the radiators on the back of each of the 12 server racks and back out into the ocean.

https://news.microsoft.com/features/under-the-sea-microsoft-...

So watercooling with seawater, a pump or two, perhaps a heat exchanger (the radiator) is involved. Server to air to pod outer surface would be way too inefficient to keep the servers operating.

Re:Microsoft's underwater data centre resurfaces after two y... by maxerickson on Monday September 14, 2020 @05:19PM (#24475279)
There's probably an internal heat exchanger that they pump seawater through. This article says they pump seawater through heat exchangers on each rack (and presumably shows the piping):

https://news.microsoft.com/features/under-the-sea-microsoft-...

Re:Microsoft's underwater data centre resurfaces after two y... by msandford on Monday September 14, 2020 @05:23PM (#24475324)
Seems kind of strange to put the whole thing underwater if they're going to pump seawater through internal heat exchangers. At that point there's not a lot of difference between putting the "submarine" underwater and leaving it out of -- but near to -- the water and pumping seawater through the same exchangers. With the added benefit of people being able to access things if something goes wrong.
Re:Microsoft's underwater data centre resurfaces after two y... by maxerickson on Monday September 14, 2020 @05:30PM (#24475385)
Their public statements so far indicate that, on the prototype, essentially nothing went wrong in a couple of years (a small number of board failures).

On the seafloor, they get constant low temperature water with ~0 external pipe and no head to pump against, so it might not obviously be better to have access.

Re:Microsoft's underwater data centre resurfaces after two y... by remarkEon on Monday September 14, 2020 @05:51PM (#24475552)
What’s different about the hardware that they’re seeing failure rates that low?
Re:Microsoft's underwater data centre resurfaces after two y... by jalk on Monday September 14, 2020 @06:07PM (#24475697)
They speculate that the removed oxygen (replaced with nitrogen) and no humans banging stuff (by accident I assume), plays a role
Re:Microsoft's underwater data centre resurfaces after two y... by Hnrobert42 on Monday September 14, 2020 @06:03PM (#24475652)
(Just speculating)

- there are issues with using water for heat exchange and that affecting flora/fauna. maybe the impact can be spread out more at the bottom of the ocean than for on shore locations

- pumping water all the way on to land and back out can take a lot of energy (water is heavy)

- coastal land is generally expensive, certainly more than seafloor which is free(?)

- coastal land is subject to storm surge in a way that deep seafloor is not

Re:Microsoft's underwater data centre resurfaces after two y... by quickthrowman on Monday September 14, 2020 @05:17PM (#24475246)
I was wondering the same thing, since it’s submerged I’m assuming they use seawater running through chillers to cool down the servers?