Microsoft declares its underwater data center test was a success

Underwater pods can reduce latency by moving cloud services closer to customers.

Read the whole story

 

[LordOfThePigs]

With enough data centres like these, all the excess heat will go straight into the sea, further contributing to climate change. The only place these data centres wouldn’t contribute further to climate change is if they were buried deep down in earth’s crust but then cooling would be an issue.

I think I prefer the idea of city center data centres where the excess heat is put to use for central heating systems

Thermodynamics fail.

Global warming is not caused by humans creating heat. There is nothing humans could do that would produce even the tiniest fraction of the thermal energy that reaches the Earth from the sun.

Global warming is caused by GHG which increase the percentage of the thermal energy from the sun which is trapped. The sun is so off the charts powerful compared to anything heat sources produced by humans that even the tiniest increase in the greenhouse effect can raise global temperatures.

This got me curious, so I did a bit of googling and some arithmetic. 2015 global energy consumption was about 575 x 10^15 BTU, or roughly 2 x 10^14 kWh. Meanwhile, averaged over the entire globe, insolation is about 6 kWh/m^2 per day, or globally about 1 x 10^18 kWh per year. If I haven't screwed anything up, that would make human energy consumption the equivalent of 0.02% of Earth's annual insolation. Personally I'd call that "within the tiniest fraction," but your point remains well taken; human energy consumption is a drop in the bucket compared to insolation.

Not to mention that a whole bunch of the energy produced by humans would have been dissipated in the environment anyway, even if we didn't use it for something useful:

- Hydro (The water would have evaporated anyway)
- Wind (The wind would have blown just the same)
- Solar (The light would have heated the ground instead)
- Nuclear Fission (the atoms would have decayed anyway)
- Geothermal (the atoms would have decayed anyway, and tidal forces would have occurred anyway)

Then only thing that is left is really fossil fuels, where for which the solar energy was stored a long time ago, and we are releasing it now, adding heat to the current atmosphere.

 

[Bongle]

- Nuclear Fission (the atoms would have decayed anyway)

This isn't quite right - The enrichment and arrangement of the fuel rods in a reactor massively accelerate the natural decay rate, since each emitted neutron gets used to usefully split another atom, rather than dissipating into nature like it did when the Uranium was in the ground. Moving forward the fissions that would've happened anyway into the present will generate a slight increase in heat in the human era.

An analogy is to think of a nuke exploding: those atoms weren't going to decay anyway when they did, they were encouraged to decay by the structure and mechanisms of the bomb.

- Solar (The light would have heated the ground instead)

Although hitting a vast field of mostly-black solar panels is going to retain more heat (on top of the 20-40% electrical energy we extract and later turn into heat through some other process) than hitting a big high-albedo chunk of desert.

 

[Statistical]

- Nuclear Fission (the atoms would have decayed anyway)

This isn't quite right - The enrichment and arrangement of the fuel rods in a reactor massively accelerate the natural decay rate, since each emitted neutron gets used to usefully split another atom, rather than dissipating into nature like it did when the Uranium was in the ground.

An analogy is to think of a nuke: those atoms weren't going to decay anyway when they did, they were encouraged to decay by the structure and mechanisms of the bomb.

That speeds up the timeframe but eventually it would have decayed away regardless. Still heat is a non issue even for fossil fuels. We have global warming because the ghg are making our greenhouse effect more efficient not because we are adding heat to the environment.

 

[AsteriaVox]

Wouldn't things like these be a great place to stack artificial reef architecture on top of? We could probably double the usefulness of things like this with some element of beneficial structure for the seafloor in areas that need it! Sounds like some teamwork could be in order, if you ask me!

 

[stevenjazz787]

Who are the people in the pic where they're posing with the pod?

 

[Cadtag]

Come on people, these are free fishing reefs / fish attractors. Super shelter habitat for groupers, and easy to reach from shore in small boats. I would want to make the containers anchor proof....

 

[Sheep Disorder]

Note that 100,000J/y is a very very small amount of energy, ~3.171 milliwatts. I don't know how much power these pods use, but I'm willing to bet it's at least a million times highter (thousands of watts)

I think I may also have messed up on my conversion of KM^3 to liters. I think it should have been ~5.6^e24.

So, 56,000,000,000,000,000,000,000,000 liters.

Let's call it 100,000 of these again. That's 56,000,000,000,000,000,000 joules you'd need, per container. Let's call it 100 billion joules per container per year. That's 56,000,000 years it would take to raise it one degree. Another way of expressing it is if you wanted to raise the ocean 1/10th of a degree, over the course of 100 years, it would take 5,600,000,000,000,000,000,000 joules per year to do so. If each unit puts out 100,000,000,000 joules per year, you would need 5,600,000,000 of these things. If each one is 10 meters squared space, you'd need 56,000,000 km^2 of space. Or approximately an area 10 times the size of Europe.

This is all if my calculations are correct. Which, again, they almost certainly are not.
In fact, my guess is that by the time I've finished editing this, they've already been proven wrong.

Edited words to make units more clear
Edited again to try fancy maths regarding number of units needed

A cubic kilometer is 1000^3 = 1 billion cubic meters. A cubic meter of water is one metric ton, 1000kg, 1000 liters.

A cubic kilometer of water is thus 1 trillion liters. That's less than one billionth of the ocean. It really is unfathomable.

I'm an engineer so I cannot do math. That's why there's Wolfram Alpha: https://www.wolframalpha.com/input/?i=volume+of+water+in+the+oceans+in+liters

If you want to be more accurate about the difference between seawater and plain water, you can also ask about that:
https://www.wolframalpha.com/input/...1+degree+celsius+difference+in+kilowatt+hours

And get an answer of 1.5 * 10^18 kWh of energy to move the temperature of the ocean by one degree.

https://www.wolframalpha.com/input/?i=worldwide+power+production gives worldwide power production as 2.2*10^13 kWh per year.

That brings us to https://www.wolframalpha.com/input/...erence+/+worldwide+power+production)+in+years for 70,000 years worth all of human electricity production to heat the oceans by 1 degree. Global warming, as we've been learning, is much faster at it than the world's largest aquarium heater could ever hope to be.

Totally irreverent because it is the local environment that matters and there are not that many suitable locations for these centers (you are not going drop these very deep anyhow). This was placed only at 117 feet (30 meters) on the sea floor which is still part of the important epipelagic zone (where photosynthesis occurs). So this will be detrimental to local environment as many species may not take a small temperature change. It is also worse when many of these and larger versions get sunk in the same area as well.

 

[Ezzy Black]

I would love to see some more information when they get done processing all the data from this.

I keep seeing people discussing the convective cooling benefits, then see all that build-up on the outside of the pod. I wonder, for instance, if that acts somewhat as an insulator, reducing that cooling over time or even if it's significant at all in that respect.

 

[Wickwick]

you also gain security benefits (nobody around to try and break in)

That is not going to be a forever thing—and I, for one, was already excited for future movies featuring SCUBA-enabled underwater physical pentests while I was writing the article.

I mean. There could legitimately be GUARD SHARKS. With frickin' lasers! (Okay, maybe not so much the lasers. And dolphins would probably be a better choice, or maybe sealions. And... and... and I still wanna see that movie.)

Given the history of both the US and Russia tapping underwater cables, I'd say the USS Jimmy Carter can expect a new addition to the MMP and a new body courtesy of the NSA's Tailored Access Operations unit in the near future.

It's a lot easier to sneak into these servers through their data connections than to need physical access.

 

[Bongle]

- Nuclear Fission (the atoms would have decayed anyway)

This isn't quite right - The enrichment and arrangement of the fuel rods in a reactor massively accelerate the natural decay rate, since each emitted neutron gets used to usefully split another atom, rather than dissipating into nature like it did when the Uranium was in the ground.

An analogy is to think of a nuke: those atoms weren't going to decay anyway when they did, they were encouraged to decay by the structure and mechanisms of the bomb.

That speeds up the timeframe but eventually it would have decayed away regardless. Still heat is a non issue even for fossil fuels. We have global warming because the ghg are making our greenhouse effect more efficient not because we are adding heat to the environment.

I'm not in any way saying that it's going to contribute to the overheating of the planet, but a nuclear reactor does increase the rate of fissions for a given hunk of U-235 over the natural background. Yes, that hunk of U-235 was going to have 6-7 half-lives in the ground and thus be mostly consumed before the sun consumes the earth, but burning even a couple percent of it instead in a couple-year stint in a reactor is a huge difference in rate, and thus human-relevant heat. Moving forward the fissions what would've happened in the far-distant future into the present makes a small, miniscule, barely calculable difference to our heat levels now. Worrying about it is like the people worrying about these data center boiling the oceans, but it is "new" heat.

And again: human waste heat isn't currently a problem on the scale of CO2-driven climate change. But saying "a nuclear reactor doesn't change how much heat the U-235 is going to emit" is being right on a technicality, because you're operating on a 4-5 billion-year timespan. The U-235 certainly wasn't going to emit the heat the reactor produced in just a few years.

Though my solar panel bit was partially in error: as long as your panel offsets electricity that would've come from a coal plant, it appears even a worst-case albedo impact can be paid back in 3 years: https://blogs.scientificamerican.com/so ... do-effect/. A new solar installation though, making new electricity and covering high-albedo land, could be problematic. Covering the Sahara in panels to power new industry might be a wash.

 

[onkeljonas]

I’m actually surprised the cooling is so seemingly easy.

Judging from the pictures the racks have pretty standard fans* moving the heat to a relatively small** volume of air, with the steel acting as a heatsink.
I can’t find figures on the temperature where the thing was placed, but a bit of searching tells me ocean floor temperatures are relatively stable around 0-3C (maybe slightly warmer in the presumably shallow area).

It uses 240kW of power, presumably the vast majority of that ends up as heat (there’s some light going in and out, not sure if that has any meaningful impact on heat).

I am not good at physics, but my initial thought was that the heat couldn’t transfer fast enough from the air to the steel - I imagined heat pipes connected to the hull, but obviously it works.
How would you go about estimating the heat that it can get rid of?

*

Spoiler

**

Spoiler

 

[Jim Salter]

you also gain security benefits (nobody around to try and break in)

That is not going to be a forever thing—and I, for one, was already excited for future movies featuring SCUBA-enabled underwater physical pentests while I was writing the article.

I mean. There could legitimately be GUARD SHARKS. With frickin' lasers! (Okay, maybe not so much the lasers. And dolphins would probably be a better choice, or maybe sealions. And... and... and I still wanna see that movie.)

Given the history of both the US and Russia tapping underwater cables, I'd say the USS Jimmy Carter can expect a new addition to the MMP and a new body courtesy of the NSA's Tailored Access Operations unit in the near future.

It's a lot easier to sneak into these servers through their data connections than to need physical access.

Everything interesting in the history of espionage comes down to "this used to be hard, but now this hard thing is easier than the thing which has been the easy thing, but nobody else noticed yet. YOINK!"

If physical security on this type of deployment gets ignored "because it's all underwater anyway", Interesting Things Will Happen. Even if it's just harvesting a pod to scavenge the metals out of it, meth-head air conditioner/catalytic converter style.

 

[Jim Salter]

I’m actually surprised the cooling is so seemingly easy.

Judging from the pictures the racks have pretty standard fans* moving the heat to a relatively small** volume of air, with the steel acting as a heatsink.
I can’t find figures on the temperature where the thing was placed, but a bit of searching tells me ocean floor temperatures are relatively stable around 0-3C (maybe slightly warmer in the presumably shallow area).

It uses 240kW of power, presumably the vast majority of that ends up as heat (there’s some light going in and out, not sure if that has any meaningful impact on heat).

I am not good at physics, but my initial thought was that the heat couldn’t transfer fast enough from the air to the steel - I imagined heat pipes connected to the hull, but obviously it works.
How would you go about estimating the heat that it can get rid of?

*

Spoiler

**

Spoiler

We haven't seen enough of this thing to know for certain that the heat dissipation is absolutely, 100%, totally passive. There might be a radiator with a low-flow pump moving cold seawater through it; you can dump IMMENSE amounts of heat that way.

There might also be a heat exchanger pumping fresh water (or mineral oil, or whatever) from heat sinks directly in the rack in a closed loop, with a passive radiator on the seawater side. AFAIK, Microsoft hasn't released full details about exactly how the cooling works, other than it's effectively free as compared to traditional refrigerative cooling.

 

[johnt007871]

On the topic of power and reliability, it occurs to me that they could potentially build flow batteries inside one of those cylinder 'pods'. So, maybe your data center looks like a lot of nearly identical looking pods - but in a ratio of 3 or 4 data pods, and one battery pod.

Scale it up to maybe 500 pods for a large datacenter - with 100 battery pods and 400 data pods, or something like that. Maybe you don't even need 1 to 4. Maybe it's like 1 to 10, as servers get more and more power efficient. And since you don't need active cooling, you save a lot of power on hvac that you don't need.

Are any of the pods used for data storage instead of compute? Seems like maybe you could build some pods are that are big enterprise storage arrays.

Would be kind of interesting to drop these near off shore wind farms, assuming they aren't too far out from the target metro area. I assume it's easier to transmit data than transport power.

 

[onkeljonas]

I’m actually surprised the cooling is so seemingly easy.

Judging from the pictures the racks have pretty standard fans* moving the heat to a relatively small** volume of air, with the steel acting as a heatsink.
I can’t find figures on the temperature where the thing was placed, but a bit of searching tells me ocean floor temperatures are relatively stable around 0-3C (maybe slightly warmer in the presumably shallow area).

It uses 240kW of power, presumably the vast majority of that ends up as heat (there’s some light going in and out, not sure if that has any meaningful impact on heat).

I am not good at physics, but my initial thought was that the heat couldn’t transfer fast enough from the air to the steel - I imagined heat pipes connected to the hull, but obviously it works.
How would you go about estimating the heat that it can get rid of?

*

Spoiler

**

Spoiler

We haven't seen enough of this thing to know for certain that the heat dissipation is absolutely, 100%, totally passive. There might be a radiator with a low-flow pump moving cold seawater through it; you can dump IMMENSE amounts of heat that way.

There might also be a heat exchanger pumping fresh water (or mineral oil, or whatever) from heat sinks directly in the rack in a closed loop, with a passive radiator on the seawater side. AFAIK, Microsoft hasn't released full details about exactly how the cooling works, other than it's effectively free as compared to traditional refrigerative cooling.

Actually, looking closer at the picture with all the racks pulled out, there's some black piping running along the top and down into the racks. Could that be a cooling loop if some kind perhaps?

 

[Wickwick]

you also gain security benefits (nobody around to try and break in)

That is not going to be a forever thing—and I, for one, was already excited for future movies featuring SCUBA-enabled underwater physical pentests while I was writing the article.

I mean. There could legitimately be GUARD SHARKS. With frickin' lasers! (Okay, maybe not so much the lasers. And dolphins would probably be a better choice, or maybe sealions. And... and... and I still wanna see that movie.)

Given the history of both the US and Russia tapping underwater cables, I'd say the USS Jimmy Carter can expect a new addition to the MMP and a new body courtesy of the NSA's Tailored Access Operations unit in the near future.

It's a lot easier to sneak into these servers through their data connections than to need physical access.

Everything interesting in the history of espionage comes down to "this used to be hard, but now this hard thing is easier than the thing which has been the easy thing, but nobody else noticed yet. YOINK!"

If physical security on this type of deployment gets ignored "because it's all underwater anyway", Interesting Things Will Happen. Even if it's just harvesting a pod to scavenge the metals out of it, meth-head air conditioner/catalytic converter style.

If you have deep sea diving technology that makes salvaging metal as you suggest an affordable option then go for it.

 

[lurch1989]

While this is great for cooling, especially up there, I can't see any other meaningful benefits to doing this.

There's so many factors which could take the thing offline, stormy weather, shipping, fishermen, leaks.

Cool idea but just doesn't seem worth the hassle

You should consider reading the article before commenting.

I did, still just seems like a heck of a lot of faff for no great benefit

 

[Azethoth666]

I, for one, am extremely excited about the convergence of physical and digital sea-based piracy.

Arrrrr. Sunken crypto treasure? Hidden google maps? Davy Jones server locker? The captains server be beefier than the rest? Have you ever been to the servers Billy?

 

[mikedelhoo]

I, for one, am extremely excited about the convergence of physical and digital sea-based piracy.

Arrrrr. Sunken crypto treasure? Hidden google maps? Davy Jones server locker? The captains server be beefier than the rest? Have you ever been to the servers Billy?

The question begs to be asked: what's a pirate's favourite chip architecture?

Spoiler

Did you really need to click? It's ARRRRRRRRM of course.

 

[yakinabe]

With enough data centres like these, all the excess heat will go straight into the sea, further contributing to climate change. The only place these data centres wouldn’t contribute further to climate change is if they were buried deep down in earth’s crust but then cooling would be an issue.

I think I prefer the idea of city center data centres where the excess heat is put to use for central heating systems

Thermodynamics fail.

Global warming is not caused by humans creating heat. There is nothing humans could do that would produce even the tiniest fraction of the thermal energy that reaches the Earth from the sun.

Global warming is caused by GHG which increase the percentage of the thermal energy from the sun which is trapped. The sun is so off the charts powerful compared to any heat sources produced by humans that even the tiniest increase in the greenhouse effect can raise global temperatures.

To wit: the Sun bombards the Earth with more energy in *ONE HOUR* than the entire human race uses in a year. One. Hour.

And there are fools who still thing solar power isn't viable....

Rain is a viable source of water too, but you don't get to chose when you get it

Which is why we store rainwater in reservoirs to use during dry seasons.

 

[pacanukeha]

that's not really a concern compared to the reduced energy usage (and therefore reduced CO2 emissions) needed for the cooling system.

There is an absolute system gain by omitting the inefficiency of what ever power system is used to cool the normal data-centre (generation/transmission/transformation/compression/cooling).
There is also a possible system gain if the power generation is non-renewable/carbon.

 

[FabiusCunctator]

If physical security on this type of deployment gets ignored "because it's all underwater anyway", Interesting Things Will Happen. Even if it's just harvesting a pod to scavenge the metals out of it, meth-head air conditioner/catalytic converter style.

I do a lot of work with traditional AM broadcasting transmission infrastructure -- the sort of sites one sees that consist of multiple towers in a field, with red obstruction lighting at night. The entire set of towers constitute a single phased-array directional antenna (the carrier frequency for this type of broadcasting is only around 1 MHz).

What's less obvious about this type of setup is that the towers only constitute half of the antenna system. The other half is the "ground system", a set of bare copper conductors (typically around 180 of them per tower), buried a few inches below the surface and extending out for about the same distance as the tower is high. It all adds up to a lot of copper sitting there near the surface.

So, the price of Cu being what it is, it has become a thing to have your transmitter site knocked off the air in the middle of the night and then discover that someone has "harvested" half of your ground system and sold it to the local scrap dealer. It's become a real problem.

So, regarding "meth-heads" , yes. Once the word gets 'round about what's inside those strange steel sausages submerged alongside the local pier, the meat-heads will appear and try to "harvest" the contents. Physical security will be a must.

 

[Entkoffeiniert]

With enough data centres like these, all the excess heat will go straight into the sea, further contributing to climate change. The only place these data centres wouldn’t contribute further to climate change is if they were buried deep down in earth’s crust but then cooling would be an issue.

I think I prefer the idea of city center data centres where the excess heat is put to use for central heating systems

Thermodynamics fail.

Global warming is not caused by humans creating heat. There is nothing humans could do that would produce even the tiniest fraction of the thermal energy that reaches the Earth from the sun.

Global warming is caused by GHG which increase the percentage of the thermal energy from the sun which is trapped. The sun is so off the charts powerful compared to any heat sources produced by humans that even the tiniest increase in the greenhouse effect can raise global temperatures.

To wit: the Sun bombards the Earth with more energy in *ONE HOUR* than the entire human race uses in a year. One. Hour.

And there are fools who still thing solar power isn't viable....

Rain is a viable source of water too, but you don't get to chose when you get it

Which is why we store rainwater in reservoirs to use during dry seasons.

Whoosh??

 

[cniru]

With enough data centres like these, all the excess heat will go straight into the sea, further contributing to climate change. The only place these data centres wouldn’t contribute further to climate change is if they were buried deep down in earth’s crust but then cooling would be an issue.

I think I prefer the idea of city center data centres where the excess heat is put to use for central heating systems

Others have covered the fact that the reduced power load from lack of AC almost certainly outpaces this, but also, it's...really unrealistic given the size of the oceans. There are over 1.3 trillion liters of seawater in the world. It takes 4,200 joules to raise it by one degree. That's 5,586,000,000,000,000 joules.

You could put 100,000 of these in the ocean, each putting off 100,000 joules a year for 100,000 years, and it would have raised the temperature of the ocean by less than 2/10ths of a degree (celsius).

You got that volume seriously wrong - Earth ocean volume is around 1.35 billion cubic kilometers , multiply that by another trilion to get the number in liters - so its 1.35e+21 liters. At 4186 J/kg/deg C that is 5.65e24 (that's 5.65 trillion trillion) Joules of energy.

edit: a name for 1e24, bbcode fix, typo
edit 2: Multi-ninja'd

 

[cniru]

Couldn't they just seal datacenters and pressurize them with nitrogen on land? That seems like that'd get you basically all the benefits of this setup without having to, you know, actually submerge the thing in the ocean.

I wonder ... MS (and others) has some very large data centers in Quincy, Washington state, which is also a farming region with high grade top soil about 20 feet thick. Just up the street is a chain of hydro dams churning out some of the cheapest power in the country. Just up a little further is Wenachee, the apple capitol of the World. There, the fruit growers use enormous warehouses filled with nitrogen to keep the fruit prior to shipping. So, you have the infrastructure and knowledge in place in Eastern Washington to do exactly what you thought, just with bits not pits (stone fruit). Ever wonder how you get a "fresh" apple in April?

Aside: these data centers also resulted in fiber to the farm in Grant county spliced off the cable that runs to Seattle and who knows where else along the Burlington Northern tracks. Friends and family there had fiber to their houses years before I got mine in Northern California.

Wenatchee is no apple capital of the World - your take is much like the "World Series" I dare say! China alone produces more apples than the rest of the world combined, and even India produces about 1/2 the US output, and we are a sub-tropical country with all its jolly pests making sub-par productivity, where Apple don't wanna grow. And lest you ask - I know Wenatchee or its claims to Apples, went to school in WSU and have close friends working at the TFREC.

The US like in many other things have outsourced their food production too, much to the chagrin of many farmers, there are too many dimensions to the current economic mess and inequality in the US.

 

[Snarky Robot]

edit 2: Multi-ninja'd

Ars: beware the roaming packs of undomesticated ninjas,

 

[Wardatrigger]

What would it be like for spots closer to the data centre though? Would the extra heat make it unpleasant for local sea life? Cooling towers on land have a noticeable effect on rivers close to warmer outflowing water.

 

[deet]

Maybe I’m wrong in assuming the more naval aspects of all this were a solved problem. The Gulf of Mexico at least is littered with seafloor vessels doing all kinds of things. I figure we know pretty well how to put tanks underwater.

For that matter, I can’t help wondering what offshore customers Microsoft might already have an eye on for these. What kind of geology can you do when your DC isn’t at the other end of a satellite link, but instead parked right under your rig?

But what’s fascinating for me is what life must be like for a server in one of these things. Five pages deep into the thread, we’re still arguing about thermodynamics and heating buildings on land (???). Heat is an interesting issue, but these aren’t underwater toasters. They’re complex systems adapted to a uniquely harsh and remote environment. I would be interested in knowing more about the design and operational considerations, and I’d love to know more about what MS learned about those aspects.

Or is all that a solved problem too?

 

[Wickwick]

Maybe I’m wrong in assuming the more naval aspects of all this were a solved problem. The Gulf of Mexico at least is littered with seafloor vessels doing all kinds of things. I figure we know pretty well how to put tanks underwater.

For that matter, I can’t help wondering what offshore customers Microsoft might already have an eye on for these. What kind of geology can you do when your DC isn’t at the other end of a satellite link, but instead parked right under your rig?

But what’s fascinating for me is what life must be like for a server in one of these things. Five pages deep into the thread, we’re still arguing about thermodynamics and heating buildings on land (???). Heat is an interesting issue, but these aren’t underwater toasters. They’re complex systems adapted to a uniquely harsh and remote environment. I would be interested in knowing more about the design and operational considerations, and I’d love to know more about what MS learned about those aspects.

Or is all that a solved problem too?

Meh. The computer doesn't know it's in a hostile environment. It actually likes the nitrogen rather than oxygenated air apparently. The question asked before was whether these cases were pressurized to match the depth or held at one atmosphere. I would be interested to know the answer to that - especially as relates to the performance of spinning rust hard drives.

As for the rest, so long as a computer in its rack receives it electricity, has its heat removed and has a data connection to the outside world, it doesn't care where it is. It's not like there's a vibration or orientation issue with being submerged.

The article describes the back-up power concern for operating off a renewable power generation system. Optical links underwater are no big thing. All the sub-oceanic data cables have repeaters along their lengths. And the heat removal was addressed partially in the article. The ocean was used as a heat sink. Personally, I'd like to know if there was a coolant loop to efficiently couple the servers to the shell or if they just relied on fans but that would couple into the operating pressure discussion as well.

 

[SixDegrees]

Maybe I’m wrong in assuming the more naval aspects of all this were a solved problem. The Gulf of Mexico at least is littered with seafloor vessels doing all kinds of things. I figure we know pretty well how to put tanks underwater.

For that matter, I can’t help wondering what offshore customers Microsoft might already have an eye on for these. What kind of geology can you do when your DC isn’t at the other end of a satellite link, but instead parked right under your rig?

But what’s fascinating for me is what life must be like for a server in one of these things. Five pages deep into the thread, we’re still arguing about thermodynamics and heating buildings on land (???). Heat is an interesting issue, but these aren’t underwater toasters. They’re complex systems adapted to a uniquely harsh and remote environment. I would be interested in knowing more about the design and operational considerations, and I’d love to know more about what MS learned about those aspects.

Or is all that a solved problem too?

Meh. The computer doesn't know it's in a hostile environment. It actually likes the nitrogen rather than oxygenated air apparently. The question asked before was whether these cases were pressurized to match the depth or held at one atmosphere. I would be interested to know the answer to that - especially as relates to the performance of spinning rust hard drives.

As for the rest, so long as a computer in its rack receives it electricity, has its heat removed and has a data connection to the outside world, it doesn't care where it is. It's not like there's a vibration or orientation issue with being submerged.

The article describes the back-up power concern for operating off a renewable power generation system. Optical links underwater are no big thing. All the sub-oceanic data cables have repeaters along their lengths. And the heat removal was addressed partially in the article. The ocean was used as a heat sink. Personally, I'd like to know if there was a coolant loop to efficiently couple the servers to the shell or if they just relied on fans but that would couple into the operating pressure discussion as well.

Another point in favor of this environment, mentioned in the article: you don't have people wandering around bumping into things, spilling coffee on things, plugging and unplugging things, and generally acting as biological sources of chaos.

My software/hardware lab would almost never have any problems if I locked all the people who use it out of it.

 

[Skyfire77]

Maybe I’m wrong in assuming the more naval aspects of all this were a solved problem. The Gulf of Mexico at least is littered with seafloor vessels doing all kinds of things. I figure we know pretty well how to put tanks underwater.

For that matter, I can’t help wondering what offshore customers Microsoft might already have an eye on for these. What kind of geology can you do when your DC isn’t at the other end of a satellite link, but instead parked right under your rig?

But what’s fascinating for me is what life must be like for a server in one of these things. Five pages deep into the thread, we’re still arguing about thermodynamics and heating buildings on land (???). Heat is an interesting issue, but these aren’t underwater toasters. They’re complex systems adapted to a uniquely harsh and remote environment. I would be interested in knowing more about the design and operational considerations, and I’d love to know more about what MS learned about those aspects.

Or is all that a solved problem too?

Meh. The computer doesn't know it's in a hostile environment. It actually likes the nitrogen rather than oxygenated air apparently. The question asked before was whether these cases were pressurized to match the depth or held at one atmosphere. I would be interested to know the answer to that - especially as relates to the performance of spinning rust hard drives.

As for the rest, so long as a computer in its rack receives it electricity, has its heat removed and has a data connection to the outside world, it doesn't care where it is. It's not like there's a vibration or orientation issue with being submerged.

The article describes the back-up power concern for operating off a renewable power generation system. Optical links underwater are no big thing. All the sub-oceanic data cables have repeaters along their lengths. And the heat removal was addressed partially in the article. The ocean was used as a heat sink. Personally, I'd like to know if there was a coolant loop to efficiently couple the servers to the shell or if they just relied on fans but that would couple into the operating pressure discussion as well.

Another point in favor of this environment, mentioned in the article: you don't have people wandering around bumping into things, spilling coffee on things, plugging and unplugging things, and generally acting as biological sources of chaos.

My software/hardware lab would almost never have any problems if I locked all the people who use it out of it.

Humans, always the weak link.

 

[dracorpg]

Keep in mind that serious datacenter operators (aka cloud providers) have been working to get their power usage efficiency (PUE = ratio of total datacenter power consumption to computer-hardware-only power consumption) as close to 1 as possible for quite some time now. Google currently reports a fleet average of 1.11 https://www.google.com/about/datacenters/efficiency/

So, compared to these highly-optimized surface datacenters, there isn't that much power to save by reducing cooling consumption to zero thanks to full seawater immersion. I mean, avoiding a ~10% overhead is always nice, but nowhere near game-changing really. The reliability increase may indeed very possibly come from running hardware sub-10°C rather than in a >20°C ambient air.

Regarding EMP resistance, clearly with that much (decently conductive) seawater above and a relatively thick steel shell these pods are well hardened against radiated effects. As for pulses conducted down the (most likely steel-armored coaxial) power cable, it's difficult to say without knowing the exact cable impedance characteristics & pulse shape... not something I have much experience with so I'll refrain from commenting further.

Edit: with underwater turbines in strong sea current areas such as here, one could indeed envision fully power-self-sufficient datacenters with only a fiber link to the shore...

 

[SixDegrees]

Keep in mind that serious datacenter operators (aka cloud providers) have been working to get their power usage efficiency (PUE = ratio of total datacenter power consumption to computer-hardware-only power consumption) as close to 1 as possible for quite some time now. Google currently reports a fleet average of 1.11 https://www.google.com/about/datacenters/efficiency/

So, compared to these highly-optimized surface datacenters, there isn't that much power to save by reducing cooling consumption to zero thanks to full seawater immersion. I mean, avoiding a ~10% overhead is always nice, but nowhere near game-changing really. The reliability increase may indeed very possibly come from running hardware sub-10°C rather than in a >20°C ambient air.

Regarding EMP resistance, clearly with that much (decently conductive) seawater above and a relatively thick steel shell these pods are well hardened against radiated effects. As for pulses conducted down the (most likely steel-armored coaxial) power cable, it's difficult to say without knowing the exact cable impedance characteristics & pulse shape... not something I have much experience with so I'll refrain on commenting further.

As noted in the article, though, those aren't major considerations. The rather stunning eight-fold improvement in failure rate, and the relatively cheap real estate that's still close to major populations, factor in a lot more heavily.

 

[Grey Bird]

With enough data centres like these, all the excess heat will go straight into the sea, further contributing to climate change. The only place these data centres wouldn’t contribute further to climate change is if they were buried deep down in earth’s crust but then cooling would be an issue.

I think I prefer the idea of city center data centres where the excess heat is put to use for central heating systems

The heat is going into the environment one way or the other anyway. Even if the heat is redirected to heating a building, the heat then goes from the building into the environment.

Also, the amount of heat the oceans absorb from these would be negligible compared to what they absorb from solar radiation every single day. You might see some minor local heating in the immediate environment, but that's not really a concern compared to the reduced energy usage (and therefore reduced CO2 emissions) needed for the cooling system.

My point about “recycling” the heat is that you don’t need to produce two units of heat (1 as a byproduct from the data center, the other for keeping buildings at liveable temperature). Byproduct heat of data center is subtracted in part from energy required to heat a building; this leading ultimately to less overall heat dumped in the atmosphere. What am I missing?

Your data center needs heat removal, and your building needs cooling during warmer months. So this doesn't work then. Also the ocean removes the heat away from the data center much more efficiently than transferring it using ducting and fans to try and heat a building. Also, if you read the article you'll see that the heat removal is only one of the problems that this solves which your idea doesn't.

 

[fl4Ksh]

Interesting idea. Solves reliability issues and provides inexpensive cooling for those power-hungry servers.

But the fiber optic cables connecting the servers to the mainland may be vulnerable to sabotage.

 

[Frennzy]

But the fiber optic cables connecting the servers to the mainland may be vulnerable to sabotage.


Replying to "Microsoft declares its underwater data center test was a success":

This is just about the dumbest possible argument against the idea. Firstly, DCs like this are part of massively geographically dispersed DCs acting in tandem to handle data and shift loading. Deploy a dozen of these around the world, and one going offline won't cause much issue. Deploy a thousand, and one going offline would be a tier 3 event..."eh, we'll fix it when we have time. Maybe". One might say it'd barely be a blip on the sonar.

Further, you do realize we have *vastly, vastly* larger undersea data cables today,right? As in, intercontinental? All over the world? Accidental damage is far more common than intentional, and yet, somehow we survive...and the internet continues to route around problems.

(oh, and it's no easier, in fact largely harder) to sabotage terrestrial cables than sub-sea cables.

 

[SixDegrees]

Interesting idea. Solves reliability issues and provides inexpensive cooling for those power-hungry servers.

But the fiber optic cables connecting the servers to the mainland may be vulnerable to sabotage.

Maybe. But they're cheap, and easy to repair. There are literal tons of undersea cables out there, and although snags occur, we know how to deal with them.

 

[Uxorious]

While this is great for cooling, especially up there, I can't see any other meaningful benefits to doing this.

There's so many factors which could take the thing offline, stormy weather, shipping, fishermen, leaks.

Cool idea but just doesn't seem worth the hassle

Why not? I mean, we already run undersea data cables across the Atlantic, which are marked as no-anchor zones anyway. Why not add a data center next to it, especially for latency reasons.

Wait till the Professor finds one of these washed up on Gilligan's Island...

 

[Nop666]

Note that 100,000J/y is a very very small amount of energy, ~3.171 milliwatts. I don't know how much power these pods use, but I'm willing to bet it's at least a million times highter (thousands of watts)

I think I may also have messed up on my conversion of KM^3 to liters. I think it should have been ~5.6^e24.

So, 56,000,000,000,000,000,000,000,000 liters.

This is the Metric system we're talking about; getting a '56' in the result means you messed something up. In fact, a cubic kilometer is a trillion (1,000,000,000,000, or 10^12) litres.



Let's call it 100,000 of these again. That's 56,000,000,000,000,000,000 joules you'd need, per container. Let's call it 100 billion joules per container per year. That's 56,000,000 years it would take to raise it one degree. Another way of expressing it is if you wanted to raise the ocean 1/10th of a degree, over the course of 100 years, it would take 5,600,000,000,000,000,000,000 joules per year to do so. If each unit puts out 100,000,000,000 joules per year, you would need 5,600,000,000 of these things. If each one is 10 meters squared space, you'd need 56,000,000 km^2 of space. Or approximately an area 10 times the size of Europe.

This is all if my calculations are correct. Which, again, they almost certainly are not.
In fact, my guess is that by the time I've finished editing this, they've already been proven wrong.

Edited words to make units more clear
Edited again to try fancy maths regarding number of units needed[/quote]

 

[Frennzy]

Wait till the Professor finds one of these washed up on Gilligan's Island...

Replying to "Microsoft declares its underwater data center test was a success":

He'd build an Xbox out of coconuts and palm leaves, so they could all play games thus alleviating the boredom of being a castaway.

Complete ignoring, of course, that he could send messages requesting rescue.