From Ars Technica, word that Microsoft is deploying pods with servers underwater.
Microsoft CEO Satya Nadella says that underwater server farms are part of the company's plans for future data centers.
Microsoft has been experimenting with underwater servers for some time.
Project Natick[*] put a server pod underwater off the coast of California in 2016. Naturally enough, the pod uses water cooling, dumping waste heat into the ocean around it. It's designed as a sealed unit, deployed for five years before being brought back up to the surface and replaced. Since then, Microsoft has deployed a larger pod off the coast of Scotland.
[*] [Natick is the name of a town in eastern Massachusetts which also happens to have a US Army Research Facility located in it. --Ed.]
The pod people are no longer people! Flash in the pan idea, or could it have some traction?
(Score: 4, Insightful) by jmorris on Saturday November 03 2018, @03:57AM (3 children)
Pros:
Lower cooling cost.
Perhaps lower real estate cost but I doubt it, any coast where you can get fiber into the water is a high cost area to operate in vs a desert somewhere a fiber is passing through.
Cons:
Everything important has to work flawlessly for five years. Individual servers can obviously fail but the cooling system can't. The power distribution system can't so much as throw a breaker. The main switch carrying Internet in can't fail. Anything goes wrong and you have to down the entire pod of servers, raise it and pay the expense of a sea mission to service it.
Bottom line:
You had better save an assload of cash on cooling to consider doing this as a business instead of a green stunt to harvest virtue. We shall see what happens after a dozen or so attempts how many make it five years without a maintenance event. If it does work it will certainly take "Lights out management" up a notch.
(Score: 3, Interesting) by edIII on Saturday November 03 2018, @04:12AM
factor in power generation from waves. There was a vertical solid-state wind generator that worked in clusters to take advantage of wind vortices. Suspended a magnetic rod in it and generated energy from the vibration. If you tie it to the ground and make them buoyant, they could rise and fall and generate energy from that.
Dunno how much gear you could power from these methods, but if you eliminated cooling costs and power costs, you just might get costs low enough profit.
Technically, lunchtime is at any moment. It's just a wave function.
(Score: 2, Touché) by Anonymous Coward on Saturday November 03 2018, @04:39AM
has to work flawlessly for five years
Heh... Microsoft
(Score: 4, Interesting) by KilroySmith on Saturday November 03 2018, @04:43AM
A lot of those issues are ameliorated if you place many pods together - say you co-locate 100 pods. You wait for 2 or 3 pods to fail, you pull the servicing ship out of the dock and replace them.
Probably have to do that someplace with a good current flowing by to avoid local underwater environmental effects from water heating...