Microsoft to dump it's computers into the sea.

[Anomalocaris]

I'm a whole lot more concerned about environmental impact than feasibility.

Yeah that was my worry as well.

Some of them are going to leak in big storms that are more likely given climate change. The sea is a corrosive and hostile environment at the best of times.

Saying that, the casing will need to be brought up again if only to replace all the computers once they become obsolete.

And saying that it will only increase the water temperature by a millionth of a degree is disingenuous because the whole point of the experiment is to see if it can scale up.

These capsules will probably not be place in the tidal zone or in unprotected depths subject to wave action.  I don’t think protecting the capsules from storms will be a challenge.   The only thing that will be exposed to storms will be the power and data cables coming up.  But we’ve had telecommunication and underwater power cables for decades.  They seldom fail due to storm.

Some may become damaged and may leak, but there is no reason why they will need to contain significant amounts toxic gas or liquid. So if they leak they flood and the computer inside is destroyed.  But nothing particularly harmful will come out.   Encapsulate electronic equipment for long duration underwater use is also very common place in telecommunication, petroleum, and naval applications and has been since 1950s.  No real serious concerns or known instances of diseaster.

Given the volume of surface water and air, our most vainglorious attempt to heat them up to any significant degree by burning fuels in our reach would amount to nothing for the foreseeable horizon.   The concern is always how human activity can indirectly cause the sun to heat the air and ocean more effectively.   Because these capsules need not be actively cooled, they will actually release less heat in total the a facility on land that must consume power to cool itself.

The total power consumed by data centers to cool electronics is much larger than one might imagine. Here in San Francisco Bay Area, data center power consumption represents up to a third of total electric load for many of the municipalities, and exceed the residential and commercial load. A single data center consumes as much power as a typical mixed city of 20000-30000 people.

[I_am_not_mafia]

Given the volume of surface water and air, our most vainglorious attempt to heat them up to any significant degree by burning fuels in our reach would amount to nothing for the foreseeable horizon.

The concern is always how human activity can indirectly cause the sun to heat the air and ocean more effectively.

Take all the data farms in the world. They produce a lot of heat. If you are heating the sea then it traps the heat on Earth for longer than if you heat the air, where it will be more quickly escape into space. If we're getting a greenhouse effect of a thin layer of methane and carbon dioxide, then we're adding to that greenhouse by first heating the oceans. Because the heat from the oceans still has to get through the atmosphere to escape the planet.

Of course data farms aren't the only contributing factor to climate change, but nor are they a negligible one either.

[The Grand Nudger]

Minor nitpick..but..electronics are among our most toxic forms of waste.  Any significant amount of them in the ocean does present a known risk.  That risk is as significant as the installation itself would be.  Dump a few boards in the bay, nbd......server farms underwater......whole different bag of worms.

They;d have to keep a pretty tight handle on extracting units for proper disposal whenever the need arose....or get some sweetheart polluter provisions.
(they;d opt for the latter, and we;d give it to them, lol)

[AFTT47]

The total power consumed by data centers to cool electronics is much larger than one might imagine.   Here in San Francisco Bay Area, data center power consumption represents up to a third of total electric load for many of the municipalities, and exceed the residential and commercial load.   A single data center consumes as much power as a typical mixed city of 20000-30000 people.

Yes, I noticed that amongst the posts here. Most people vastly underestimate the heat problem and just how much energy it takes to deal with it. Engineers are not idiots. It's guaranteed they have done the math before concluding this approach is worth a look.

[Anomalocaris]

Minor nitpick..but..electronics are among our most toxic forms of waste.  Any significant amount of them in the ocean does present a known risk.  That risk is as significant as the installation itself would be.  Dump a few boards in the bay, nbd......server farms underwater......whole different bag of worms.

They;d have to keep a pretty tight handle on extracting units for proper disposal whenever the need arose....or get some sweetheart polluter provisions.
(they;d opt for the latter, and we;d give it to them, lol)

That is true, but unless the capsule is totally smashed, as might happen if a big ship dropped an anchor on it or ran aground on top of it, the solid toxic material will remain contained inside the damaged hull of the capsule, and the rate of leakage of toxic material out of the hill will be very slow, giving plenty of time to pull the damaged capsule out of the ocean. It is liquid or gas waste that will likely quickly leak out through minor or moderate damage to the capsule Hull.

[johan]

No need for space grade components. Weight won't be a factor like it is with objects bound for space so you can pack lots of extra backup units in there. It won't be Best Buy cheap but you should be able to do it for way less than the cost of space grade. My point was creating computer systems that work reliably hands off for extended periods of time is hardly uncharted territory.

True, to the extent that they have the room for massive redundancy. I still think, like Anomolacaris, why can't they be retrieved? That seems seriously not thought out.

I'm sure they can be retrieved. But there's going to be a cost associated with that. So if you've got to send the boat and divers out every 18 months, its probably not cost effective. However if you design them to be able to deploy and run for 5 or 10 years with no physical maintenance, then you can probably make the costs work out. You deploy them, leave them on the bottom for X years, pull them up, gut them, install new machines and put them back on the bottom for another X years. In the meantime you've saved gobs of money by having no measurable cooling costs. You'd also see savings from not paying any property taxes, not having to maintain any landscaping or parking lots etc.

[The Grand Nudger]

So long as you weren;t hit with a remediation fund obligation.