Slashdot Mirror


Microsoft Sinks Data Centre Off Orkney To Test Energy Efficiency (bbc.co.uk)

An anonymous reader writes: Microsoft has sunk a data centre in the sea off Orkney to investigate whether it can boost energy efficiency. The data centre, a white cylinder containing computers, could sit on the sea floor for up to five years. An undersea cable brings the data centre power and takes its data to the shore and the wider internet -- but if the computers onboard break, they cannot be repaired. The operation to sink the Orkney data centre has been an expensive multinational affair. The cylinder was built in France by a shipbuilding company, Naval, loaded with its servers and then sailed from Brittany to Stromness in Orkney. There, another partner, the European Marine Energy Centre (EMEC), provided help including the undersea cable linking the centre to the shore. "This is a crazy experiment that I hope will turn into reality" said Ben Cutler, who is in charge of what Microsoft has dubbed Project Natick. "But this is a research project right now -- and one reason we do different types of research into data centres is to learn what makes sense before we decide to take it to a larger scale."

155 comments

  1. A lot of sunken capital by postbigbang · · Score: 1

    Let the puns begin.

    An idea that's all wet?

    Swimming with the fishes, Microsoft?

    Offshoring your data centre?

    --
    ---- Teach Peace. It's Cheaper Than War.
    1. Re:A lot of sunken capital by Tablizer · · Score: 2

      It's missing a key device diver.

    2. Re:A lot of sunken capital by tlhIngan · · Score: 2

      Shark Week's coming soon!
      Azure vs. the Great White
      Can Megalodon do Office365?
      Teaching Sharks to Surf (the Internet)

    3. Re:A lot of sunken capital by Anonymous Coward · · Score: 0

      Brings new meaning to: Boil the ocean.

      >rim shot

    4. Re:A lot of sunken capital by Thud457 · · Score: 1

      That's your solution to everything. Under the sea, its not going to happen.

      --

      the preceding comment is my own and in no way reflects the opinion of the Joint Chiefs of Staff

    5. Re:A lot of sunken capital by Anonymous Coward · · Score: 0

      Let the puns begin.

      An idea that's all wet?

      Swimming with the fishes, Microsoft?

      Offshoring your data centre?

      Dunno, about any of that. What I do know is that this datacenter is going 'float-about' the next time the N-Atlantic gets into a bad mood like a number of similar experiments have done in that region in years past.

    6. Re:A lot of sunken capital by sjames · · Score: 3, Funny

      Microsoft has been sinking data centers since NT.

    7. Re:A lot of sunken capital by Anonymous Coward · · Score: 0

      Not with that attitude.

  2. Watercooling by mi · · Score: 1

    Let the puns begin.

    It is all about watercooling, man...

    --
    In Soviet Washington the swamp drains you.
    1. Re:Watercooling by cellocgw · · Score: 2

      It is all about watercooling, man...

      So, since they're cooling PC boards, is it ....

      waterboarding?

      --
      https://app.box.com/WitthoftResume Code: https://github.com/cellocgw
    2. Re:Watercooling by Joce640k · · Score: 3, Insightful

      Never mind the puns, how long before somebody steals it?

      A bunch of Somalian fishermen with a supply of large inflatable bags will have that thing off the sea floor in no time.

      --
      No sig today...
    3. Re:Watercooling by TheRaven64 · · Score: 1

      It looks pretty heavy. It's also not very far out to sea, so I doubt you could steal it without anyone seeing and it wouldn't take MS very long to notice an entire datacentre dropping off the Internet and sending someone out to take a look...

      --
      I am TheRaven on Soylent News
    4. Re:Watercooling by MoarSauce123 · · Score: 1

      All it needs is an anchor dragged across the cable. I hope this flops, the oceans are already heating up enough, no need to accelerate this.

    5. Re:Watercooling by MoarSauce123 · · Score: 1

      Well yea, but by the time they take a look it is gone. Maybe they put a battery backed up GPS and a pressure sensor in the can. That will tell them immediately if the can moves more than expected.

    6. Re:Watercooling by Cederic · · Score: 2

      You must be American. Somalia is a 9000km boat ride from Orkney, assuming they use the Suez canal and don't get arrested or sunk on the way.

      I'm thinking there are easier ways for them to earn a living.

      Shit fuel costs and canal fees will be higher than the value of whatever they dredge up.

    7. Re:Watercooling by TheRaven64 · · Score: 1

      I doubt you can get it very far in the time that it takes to send a security guard or the police out. The thing is several tons and you can't just drag it through the water, you need to lift it up onto a boat, during which point you'll trigger all sorts of alarms (vibration, motion, shock).

      --
      I am TheRaven on Soylent News
    8. Re:Watercooling by Anonymous Coward · · Score: 0

      It is like you don't comprehend just how big the oceans on this planet of ours are. The heating effect of this, or even a million of these will be insignificant on a global scale. There will quite possibly be some local effects though with some organisms taking advantage of the localised heating in some way, but likely nothing to be concerned about.

  3. Even better by Big+Bipper · · Score: 5, Interesting

    Why not build data centers into the bases of offshore wind turbines. You would still get the, for all practical purposes, infinite heat sink of the ocean or large lake, cheap energy ( most of the time when the wind blows ), access for repair, and the data cable could be laid with the power cable from the turbine. Everybody wins.

    --
    You live and learn, or you don't learn much.
    1. Re: Even better by itamblyn · · Score: 1

      Good point!!!

    2. Re:Even better by Anonymous Coward · · Score: 0

      Nice!

    3. Re:Even better by Anonymous Coward · · Score: 0

      Cool!!

    4. Re: Even better by Anonymous Coward · · Score: 0

      Replacing a failed HDD is going to be a real bitch? Though not nearly as bad as one sunk several fathoms deep.

      Anyone up for a dive?

    5. Re:Even better by Anonymous Coward · · Score: 1, Funny

      plus they can use the fan for cooling

    6. Re:Even better by Solandri · · Score: 1

      If you read Microsoft's news article on this, the point of this project isn't to cool a datacenter using the ocean as a heat sink. It's to build a self-contained datacenter which can be customized to order, is easily shipped (fits on the back of a truck), then deployed at the bottom of the ocean where it operates for years without human intervention. All you need to do is plug in the power and data cables to something onshore.

      I suspect putting it at the bottom of the ocean is more about preventing people from breaking in and stealing the equipment inside, rather than cooling it. It's got a built-in self-destruct mechanism which triggers if anyone tries to break in while it's on the seabed, thus eliminating any incentive for thieves. This goes along with the self-contained theme (no need to hire security guards).

    7. Re: Even better by jdschulteis · · Score: 1

      Replacing a failed HDD is going to be a real bitch? Though not nearly as bad as one sunk several fathoms deep.

      Anyone up for a dive?

      The economics might well work out such that failed hardware wouldn't be replaced, just taken offline. The combined wind turbine and data center would produce slightly more power and slightly less computation.

    8. Re:Even better by whoever57 · · Score: 2

      the point of this project isn't to cool a datacenter using the ocean as a heat sink. It's to build a self-contained datacenter which can be customized to order, is easily shipped (fits on the back of a truck), then deployed at the bottom of the ocean where it operates for years without human intervention.

      So why put it at the bottom of the ocean? Why not in a container near the ocean? In most of the places where this will be deployed, land is cheap. The Microsoft document you point to isn't very clear on the advantages, but it does include this:
      "The world's oceans at depth are consistently cold, offering ready and free access to cooling, which is one of the biggest costs for land-based datacenters."
      It does also mention the idea of co-locating with a turbine, but I think the statement above suggests that energy efficiency is one of the primary goals.

      --
      The real "Libtards" are the Libertarians!
    9. Re:Even better by jdschulteis · · Score: 1

      I think saving money on cooling is the primary motivation behind the project, with security an additional consideration. From the Microsoft page that you linked:

      In fact, Naval Group 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. Findings from phase 1 of Project Natick indicate water from the datacenter rapidly mixes and dissipates in the surrounding currents.

      From the IEEE Spectrum article about Project Natick that Microsoft links to:

      When Sean James, who works on data-center technology for Microsoft, suggested that the company put server farms entirely underwater, his colleagues were a bit dubious. But for James, who had earlier served on board a submarine for the U.S. Navy, submerging whole data centers beneath the waves made perfect sense.

      This tactic, he argued, would not only limit the cost of cooling the machines—an enormous expense for many data-center operators—but it could also reduce construction costs, make it easier to power these facilities with renewable energy, and even improve their performance.

      How about security? Is your data safe from cyber or physical theft if it’s underwater? Absolutely. A Natick site would provide the same encryption and other security guarantees of a land-based Microsoft data center. While no people would be physically present, sensors would give a Natick pod an excellent awareness of its surroundings, including the presence of any unexpected visitors.

    10. Re:Even better by Anonymous Coward · · Score: 0

      These seems like a bad idea. Did anyone on this project ever run a reef tank? The ocean is FILLED with stuff that loves to cling to anything it can get a grip on and start building calcified deposits. Radiators, even with massive water flow, will be super hard to keep operating efficiently long-term in this situation. And those pumps will be crusting up in no time, even at depths beyond the sun's reach.

      Also: Queue the bitching about warming the oceans, thus increasing the overall heat load on the earth.

    11. Re: Even better by mysidia · · Score: 2

      Replacing a failed HDD is going to be a real bitch?

      At our scale, messing with individual HDD's isn't economical... we'll just replace the entire datacenter when necessary.

      or... The admin will enter a command to blow electronic fuses on the failed HDD's data and power lines to ensure it fully disconnects from the backplane, then copy and paste a new License Key received from the storage manufacturer to activate one of the pre-stocked replacement bays. Either that, or the storage will be All Solid-State with an automatic failover system and artificially reduced capacity to provide increased redundancy. Ship it with 4X the modules required for the unit to work over its service life within the expected failure rates, and there will be no need to replace anything.

    12. Re:Even better by drinkypoo · · Score: 1

      Why not build data centers into the bases of offshore wind turbines.

      Because wind turbines are regularly struck by lightning.

      --
      "You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
    13. Re:Even better by Anonymous Coward · · Score: 0

      These seems like a bad idea. Did anyone on this project ever run a reef tank? The ocean is FILLED with stuff that loves to cling to anything it can get a grip on and start building calcified deposits. Radiators, even with massive water flow, will be super hard to keep operating efficiently long-term in this situation. And those pumps will be crusting up in no time, even at depths beyond the sun's reach.

      Also: Queue the bitching about warming the oceans, thus increasing the overall heat load on the earth.

      I was never stationed on a submarine, just a surface ship, but what you say about the heat exchangers is absolutely true. They require a lot of regular maintenance. I remember one time we pulled a small crab out of the filter basket during monthly maintenance and let it crawl all around the room for a while before we chucked it over the side.

      Perhaps they have some amazingly better system in mind than what we used, but the theory was certainly the same.

    14. Re: Even better by TheRaven64 · · Score: 1

      Google stopped doing that over a decade ago and I presume Azure has a similar policy. A disk fails, they offline the machine. Enough machines in a rack fail, and they replace the rack. This datacentre is specifically designed not to be accessible by humans (the gas inside has had oxygen and water vapour removed, to reduce the damage that these do to components) and is operating under the same model: you keep using it until enough hardware is dead that it's worth pulling it up and doing a complete refit.

      --
      I am TheRaven on Soylent News
    15. Re:Even better by thegarbz · · Score: 1

      Huh? So? That's a very minor engineering problem compared to most of the other issues.

    16. Re:Even better by MoarSauce123 · · Score: 1

      That is too European of an idea for a US company.

    17. Re:Even better by drinkypoo · · Score: 2

      Huh? So? That's a very minor engineering problem compared to most of the other issues.

      It really isn't. What happens when 1.21 GW passes through your server while the HDD spins up past 8,800 RPM?

      --
      "You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
    18. Re:Even better by thegarbz · · Score: 1

      The minor engineering problem prevents the 1.21GW passing through your server. Seriously this is especially trivial to do in this scenario with easy energy dissipation, easy shielding, local earth reference due to local generation, and isolation of external connections (unless you don't chose to connect via fibre because you don't like bandwidth in your datacentre).

      This is literally one of the best cases for lightning protection. It is far worse putting a datacentre in a building on land.

    19. Re:Even better by Anonymous Coward · · Score: 0

      I suspect putting it at the bottom of the ocean is more about preventing people from breaking in and stealing the equipment inside, rather than cooling it. It's got a built-in self-destruct mechanism which triggers if anyone tries to break in while it's on the seabed, thus eliminating any incentive for thieves. This goes along with the self-contained theme (no need to hire security guards).

      Neither TFA or the article you linked mention anything about security, much less a self destruct mechanism. I would not worry about people trying to break in and steal the hardware, but about sabotage. All someone needs to do is punch a hole in it, or cut the cable and the whole datacenter is worthless.

    20. Re:Even better by Anonymous Coward · · Score: 0

      whoosh.

      That's the sound of the data center traveling 30 years back in time.

  4. Not a long term solution. by Going_Digital · · Score: 2, Insightful

    This is clearly not a long term solution, the oceans are warming and that is already causing concerns. Sticking a bunch of immersion heaters in the ocean is not exactly going to help.

    1. Re:Not a long term solution. by Vreejack · · Score: 1

      It's not going to hurt, either. The amount of heat generated by all the world's servers would be undetectable in the ocean.

      --
      "Will future ages believe that such stupid bigotry ever existed!" -- Ivanhoe
    2. Re:Not a long term solution. by Oswald+McWeany · · Score: 2, Insightful

      This is clearly not a long term solution, the oceans are warming and that is already causing concerns. Sticking a bunch of immersion heaters in the ocean is not exactly going to help.

      Against the vastness of the ocean; underwater data centers are going to make no statistical difference to the temperature of the ocean.
      Even if temperatures in the environment raise by the forecasted 2C- that's not going to drastically impact the cooling ability of the ocean either.

      Of course, it would be even better if the data center was in low orbit.

      --
      "That's the way to do it" - Punch
    3. Re:Not a long term solution. by MightyYar · · Score: 2

      Alright, I need help here. Sarcasm or not?

      --
      W..w..W - Willy Waterloo washes Warren Wiggins who is washing Waldo Woo.
    4. Re: Not a long term solution. by Anonymous Coward · · Score: 1

      Actually the vacuum of space is an insulator, it's quite difficult to cool stuff in orbit. More precisely, the minimum achievable temperature is much lower, but the speed to achieve it is also slower.

    5. Re:Not a long term solution. by Anonymous Coward · · Score: 0

      Trolling much? If anything it will reduce the impact on global warming due to the reduction in power use from reduced cooling needs (assuming that power comes from fossil fuels).

    6. Re:Not a long term solution. by Anonymous Coward · · Score: 1

      The data centers would be *harder* to cool in low orbit, so that wouldn't make sense for the data-center builders, minus the abundant solar energy.

    7. Re:Not a long term solution. by godrik · · Score: 1

      This is clearly not a long term solution, the oceans are warming and that is already causing concerns. Sticking a bunch of immersion heaters in the ocean is not exactly going to help.

      Well. It is that, or run a bunch of AC systems to cool the machines down. This solutions saves about half the energy, so I say, it is still a win. (Albeit not the perfect solution.)

    8. Re:Not a long term solution. by dj245 · · Score: 1

      This is clearly not a long term solution, the oceans are warming and that is already causing concerns. Sticking a bunch of immersion heaters in the ocean is not exactly going to help.

      Against the vastness of the ocean; underwater data centers are going to make no statistical difference to the temperature of the ocean. Even if temperatures in the environment raise by the forecasted 2C- that's not going to drastically impact the cooling ability of the ocean either.

      Of course, it would be even better if the data center was in low orbit.

      A bigger point is that that heat would eventually be dispersed across the world anyway.

      Current best environmental practice is to use air cooled heat exchangers, since everything else has been restricted. That is what power stations and datacenters are mostly doing these days.

      It is more efficient to simply use seawater or river water at land installations as a heatsink to dump the AC rejected heat into. This was widely used in the 1960s and early 1970s. However, national/local environmental agencies put up regulations against this since you may disrupt the local environment. Therefore a very long pipe is needed, which periodically must be cleaned since marine life loves to attach itself to anything in the water.

      The next best thing is to create an artificial lake- that used to be how power plants in the late 70s and 80s did things. But again, this is now frowned upon due to water consumption. Wet parabolic cooling towers were a good thing for a while, until the water usage environmental permits made those impractical as well. Now everybody uses air-cooled heat exchangers, which are less efficient than any of these options. I think people underestimate how much regulation has been put in place since the 1960s even for something as mundane as dumping excess heat.

      Ultimately any energy savings to be had by sinking datacenters probably isn't worth the added infrastructure cost. Adding "marine" to anything tends to make the cost 3x what it would otherwise.

      --
      Even those who arrange and design shrubberies are under considerable economic stress at this period in history.
    9. Re:Not a long term solution. by Anonymous Coward · · Score: 1

      Back of envelope calculation gives me (assuming global data centre usage remains constant) an ocean temp rise of 2.62e-7 Kelvin/year. Not significant.

    10. Re:Not a long term solution. by cellocgw · · Score: 2

      This is clearly not a long term solution, the oceans are warming and that is already causing concerns. Sticking a bunch of immersion heaters in the ocean is not exactly going to help.

      Even worse, what if they build one of these near R'lyeh ? He might wake up from the warmth.

      --
      https://app.box.com/WitthoftResume Code: https://github.com/cellocgw
    11. Re: Not a long term solution. by Anonymous Coward · · Score: 1

      Yup, to explain what he means better:

      There are three types of heat transfer in most of everyday life; conductive (air blowing across something or another object touching, pulling the heat away), convective (heat rises due to decreased density), and radiant (IR radiation, heat lamps, the heat you feel at a distance from a camp fire). Without an atmosphere around a spaceship you loose two of the three types and must depend on IR radiation alone. Underwater you still get conductive and convective, and conductive heat transfer to water is much much better than to air.

    12. Re:Not a long term solution. by barc0001 · · Score: 2

      The warming effect of these "immersion heaters" on the oceans is insignificant even if all datacenters in the world were already sunk in the ocean. Every second of every day, the Earth is hit by 1.74 x 10^17 watts of energy from the sun. So even 100 gigawatts of "immersion heaters" in the ocean would be less than 0.001% of the heating effect that the oceans already get from solar radiation.

      The real benefit is that 15-20% reduction in electricity use for cooling, especially if that's 15-20% less fossil fuel generated electricity, which contributes to greenhouse gases.

    13. Re:Not a long term solution. by Memnos · · Score: 2

      Ultimately any energy savings to be had by sinking datacenters probably isn't worth the added infrastructure cost. Adding "marine" to anything tends to make the cost 3x what it would otherwise.

      They could offset the extra cost with extra "value" - instead of just adding "marine" they could add "marine blockchain".

      --
      I don't trust atoms -- they make up stuff.
    14. Re:Not a long term solution. by Anonymous Coward · · Score: 1

      It's not going to hurt, either. The amount of heat generated by all the world's servers would be undetectable in the ocean.

      While you'd never really need to worry about heating the ocean, creating a warm spot (or even a convenient surface) in the ocean might cause new problems ... because the things which live in the ocean will be attracted to it as a place to live.

      Your submerged data centre will become an artificial reef pretty quickly.

      What are you doing today, Bob?

      Oh, you know, it's Wednesday, so I have to go scrape the barnacles off of the data centre's cooling fins.

      Well, that, and discovering whatever the bacteria and fungus has evolved into in your little low oxygen cave when you open it -- some creepy man-eating data centre goo.

    15. Re:Not a long term solution. by cyn1c77 · · Score: 1

      This is clearly not a long term solution, the oceans are warming and that is already causing concerns. Sticking a bunch of immersion heaters in the ocean is not exactly going to help.

      You're totally right! We should put them in the arctic and antarctic instead! It's colder there!

    16. Re:Not a long term solution. by Anonymous Coward · · Score: 0

      Which is nothing to say of all of the thermal power stations along the coastlines of the continents that have been using seawater for cooling for decades....

    17. Re:Not a long term solution. by Anonymous Coward · · Score: 0

      Heat from TODAY'S servers would be undetectable in the ocean. But remember how humans react when a desirable resource seems limitless and free: they take only a few decades to run up against the limits.

      Think forests, passenger pigeons, cod, petroleum, clean air...it always happens, we increase our ability to use the resource exponentially and then a few years later, it's going, going, gone.

      Imagine automated assembly and deployment sinking millions of these datacenter pods, cranking solar and wind power through cables into what are really just large-scale heating elements. Is there a limit to humanity's appetite for computation? No.

    18. Re: Not a long term solution. by Anonymous Coward · · Score: 0

      So... imagine a Beowulf cluster of these?

    19. Re:Not a long term solution. by TheRaven64 · · Score: 1

      I wondered about this. If the heat were spread throughout the ocean, it would be a negligible impact, but raising the temperature in a bay by one degree can have quite significant ecological impacts. It depends a lot on how much tidal flow there is and how much heat the thing is dissipating.

      --
      I am TheRaven on Soylent News
    20. Re:Not a long term solution. by Anonymous Coward · · Score: 0

      A bay probably isn't a good location both for the reasons you give and the fact that there are likely to be lots of ship anchors dropped in a bay which could damage the data and power lines. In the open ocean, there will be some very localised heating, by enough water movement for it not to be anything more than that.

  5. Securing physical access? by sinij · · Score: 1

    I hope it takes more than just some scuba gear to get physical access to this data center.

    1. Re:Securing physical access? by Oswald+McWeany · · Score: 1

      I hope it takes more than just some scuba gear to get physical access to this data center.

      Don't worry, the Russian have already tapped the data line leading to the data center. They can give us a backup if someone sabotages the data center.

      --
      "That's the way to do it" - Punch
    2. Re: Securing physical access? by Anonymous Coward · · Score: 0

      You probably have to hoist it out of the water first, unless you're ok with a bunch of electronics full of seawater.

    3. Re:Securing physical access? by Anonymous Coward · · Score: 0

      sharks. sharks with frickin' lasers....

    4. Re: Securing physical access? by Anonymous Coward · · Score: 0

      I'm sure the next James Bond or Mission: Impossible movie will have a plot involving inflating a large underwater balloon around the entire center or somesuch.

  6. Sysads wanted - mermaids encourage to apply by Anonymous Coward · · Score: 0

    Microsoft will be at the forefront of inclusion and diversity!

    1. Re:Sysads wanted - mermaids encourage to apply by Anonymous Coward · · Score: 0

      No need for mermaids, you can just submerge suspended by a neckbeard regular sysadmins to work on this.

  7. So... by LordHighExecutioner · · Score: 1

    ...instead of stealing just the data, hackers will steal the whole data center ?

    1. Re:So... by Anonymous Coward · · Score: 0

      ...instead of stealing just the data, pirates will steal the whole data center ?

      FTFY. After all, this is on the high-seas. ;)

  8. Repairs by darkain · · Score: 4, Interesting

    Servers in data centers now are hardly ever repaired. Why spend the money? When you're running 10,000 servers and 1 breaks? What is the cost of that single unit vs the time to troubleshoot and solve the issue? All of the software and data is designed to be redundant anyways nowadays. The data will just be shifted around, and the processing load shifted as well. So having no access to fix things is mostly a moot issue. And 5 years? Thats about the max length of a server in a data center as it stands right now as it is. Overall, this sounds like a good scenario!

    1. Re:Repairs by Anonymous Coward · · Score: 5, Interesting

      Servers in data centers now are hardly ever repaired. Why spend the money? When you're running 10,000 servers and 1 breaks? What is the cost of that single unit vs the time to troubleshoot and solve the issue?

      Do you have any idea how cheap it is to hire a warm body, even if they got the super duper awesome A+ certification, to maintain thousands and thousands of servers? At a university, we pay ours $11/hour for less than ten hours per week to service a thousand servers. He pays for himself many times over every week compared to letting even a single server go unrepaired.

      Let's say you can't get labor that cheap and your version of cheap labor costs $30/hour with benefits (keeping in mind that this is just slightly more than a menial job to diagnose and replace DIMMs, motherboards, hard drives, etc so $30 is too much IMO). For 1000 servers you probably need much less than ten hours in a typical environment, but let's round up to ten. That's $300/week. If he fixes just one $5000 server per week, he has paid for himself even if he's on Facebook during times when nothing is broken. We are an HPC center and stress these servers to the max, so we do have multiple parts failures in a typical week.

      What morons just let servers die and throw them away when all you need is a replacement DIMM? I realize that some people are stupid or just haven't done that math but wow, that sounds idiotic to just throw away servers when it's *much* cheaper to fix them. (And yes, I realize that having to dive down to the data center wouldn't be a cheap or viable solution, so my math doesn't apply in the case described by this article).

    2. Re:Repairs by Asgard · · Score: 2

      How many more machines could you fit in a given space if you didn't have to handle a human gaining non-disruptive access to a particular machine? At scale people start looking at a rack as the unit-of-replacement, maybe even a row. In this scenario the entire sinkable unit becomes the unit of failure.

      This isn't particularly new, Sun had Project BlackBox https://en.wikipedia.org/wiki/... in 2008.

    3. Re:Repairs by ctilsie242 · · Score: 1

      In my experience, a DC doesn't need that much once it is up and running. You need facilities people (HVAC, power, security, etc.), but that is for any building. For the servers themselves, you really just need a couple operators, at worst maybe on a 24 hour shift, but even 9-5 would be just fine.

      With components like HDDs being replaced by SSDs that have a significantly larger MTBF, it really doesn't take that many people to man a DC.

    4. Re:Repairs by Anonymous Coward · · Score: 0

      In 5 years, the entire data center will rapidly lose use because you have no access to upgrade any of the units. Low performing servers (by those future standards) will slowly begin to fail until none operate.

    5. Re:Repairs by sloth+jr · · Score: 1

      One thing that can have more catastrophic problems is failure of network gear. Usually this is some of the more reliable gear in a DC, but failures certainly aren't unheard of (and trouble in cable and transceivers is a lot more common than switch gear failure - though usually the failure footprint is a lot smaller). Though every DC work its salt will have redundant network paths, that still might mean you'd be running without redundancy for the duration of the installation. That will definitely be a problem during network gear upgrades.

      Still, I like the overall concept, but wanted a few more details - eg, how the heat exchangers work, if this is going to be oxygen-limited, does that mean the cylinder will be nitrogen filled? Is there any provision for dealing with onboard leakage (bilge pumps)? Batteries on-board?

      Finally, I know this is just a research project, but I'd like to see a cost breakdown compared to a standard colo build.

    6. Re:Repairs by Anonymous Coward · · Score: 1

      What morons just let servers die and throw them away when all you need is a replacement DIMM? I realize that some people are stupid or just haven't done that math but wow, that sounds idiotic to just throw away servers when it's *much* cheaper to fix them.

      That DIMM and its motherboard are both going to be replaced soon anyway, by hardware that is more efficient (that is, makes better use of the existing power and cooling infrastructure). So, the repaired server is not worth as much as a new server.

      There is also a cost in maintaining an inventory of replacement parts.

      I know some laptops have dispensed with DIMMs in favor of RAM soldered to the motherboard, I'm sure someone has done the math as to whether that makes sense for datacenter servers.

    7. Re:Repairs by Anonymous Coward · · Score: 0

      You forget the impact on the environment. Throwing away a server because a DIMM or CPU failed? How much more environmentally unfriendly can you get?

    8. Re:Repairs by mysidia · · Score: 1

      trouble in cable and transceivers is a lot more common than switch gear failure

      When each cable enters the datacenter --- you could have a "redundancy pigtail" spliced onto each incoming fibre lead that connects each strand of fibre to three or four active MUX units (or DWDM modules) that allow for several different connections using different light frequency ranges.

      Connect the servers to a switching matrix that users MUX units as well to mate the optical connections between servers and switches, and then you can tolerate the failure of entire switches without losing any network ports on your servers.

      If a transceiver goes bad.... just turn off the failed transceiver and adjust the physical light path on the DWDM Mux matrix to attach that network cable to a spare connection on the same switch or spare switch that has a working transceiver.

    9. Re:Repairs by angel'o'sphere · · Score: 2

      I realize that some people are stupid or just haven't done that math but wow, that sounds idiotic to just throw away servers when it's *much* cheaper to fix them.
      Plenty of big data centers do that.
      Google uses to have cargo containers fitted out with racks of computers, drives and UPSs.
      They placed them somewhere and let them run, they never repaired anything, when the whole container dropped below 50% capacity they dismantled it and threw it away (because that took 3 or 4 years and the then actual tech was more advanced)
      No idea if they still follow that policy.

      --
      Cost free eBook I read (by iBook/Kobo/Amazon/ObookO/Gutenberg etc.): "The Green Odyssey" by Philip Jose Farmer.
    10. Re:Repairs by Anonymous Coward · · Score: 0

      absolute bullshit. warm bodies are expensive. Microsoft and many of the other large cloud vendors no longer waste money and time for that sort of troubleshooting. Servers go offline in the rack and then are collected and replaced weekly or monthly, depending on the age they are either disposed of or scavenged for parts or if they are one of the unusually expensive models they "might" be repairs. As you scale up avoiding humans in the process is essential, a human also requires workspace, clearance checks, management and monitoring etc etc, you currently ignore all those costs but as you scale up they quickly add up.

    11. Re:Repairs by Anonymous Coward · · Score: 0

      all true, but in a conventional data centre when the components are obsolete you wander in with a new one and "five minutes" later have upgraded a few servers. Not quite as easy when you have to lift it to the surface

    12. Re:Repairs by Anonymous Coward · · Score: 0

      and what does the work area you have set aside to repair these servers cost? how about the inventory of replacement parts that may or may not ever be used? how about the added insurance costs for having a warm body their? also remember a used server only has a fraction of the value of a new one, so unless it was a brand new server put in the rack today then it doesn't have a value of $5000. What about the security and HR processes for that person, how much do they cost you? Never mind I am sure your maths is right and Googles, Microsoft's, Amazons etc maths is just wrong and they don't understand how to do Datacentres at scale.

    13. Re:Repairs by gravewax · · Score: 1

      The fact you think the cost of someone to do this is only the hourly rate means you are someone that shouldn't be making these decisions as you don't understand the costs.

    14. Re:Repairs by Cederic · · Score: 1

      None at all for this one.

      It's about as dark site as you'll get. Run until it's broken, then recycle the whole data centre.

    15. Re:Repairs by Anonymous Coward · · Score: 0

      Moore's law has "stopped" such that this calculation may change slightly. I put quotes since improvements that happened in 18 monthes or 2 years happen in 3 or 4 years.
      Soldered RAM in servers? maybe if you're designing your own server boards, but on laptops the main draw is to use LPDDR3 (eventually LPDDR4 or LPDDR5) or in other words smartphone memory. They should make registered ECC LPDDR for that to work, i.e. the tech doesn't exist yet. Or you can solder regular DDR4 chips but there's less of a saving here if any. Those DIMMs have a vertical dimension to store the chips. So let's make stacked DDR4 or DDR5 dies but this will cost money.

    16. Re:Repairs by Anonymous Coward · · Score: 0

      I'm a farmer. When a crop fails I salt the earth and bury toxic and nuclear waste there for good measure. I also killed a few wives who had a miscarriage.

    17. Re:Repairs by cmseagle · · Score: 1

      When you pull out the bad rack/row and replace it with a new one, do you throw it straight into the dumpster? Or, do you pay OP's $30/hr warm body to repair it so that it can be slotted in as the replacement for the next rack/row that fails?

      "Rack as the unit-of-replacement" necessarily implies that you're just throwing away the hardware after it's replaced. That only makes sense if abandoning repairability gives some other desirable trait (lower initial cost, better performance, lower cost of operation, smaller physical footprint). This is an extreme case where Microsoft has totally abandoned repairability in exchange for lower cooling costs.

  9. In the Fine Article... by Anonymous Coward · · Score: 0

    A Microsoft spokesman was quoted as stating that the servers will not be running any version
    of Windows. When pressed as to why, the spokesman indicated reliability as the key factor;

    "we take this operation seriously - we're not going to torpedo
    this experiment with something crazy like Windows 10 Enterprise."

    Asked weather Bill was going to be included in the experiment, the Microsoft spokesman
    briefly smiled and said "no comment."

    CAP === 'abandons'

  10. Nitrogen by Vreejack · · Score: 1

    If eliminating oxygen and water from the air are so important, it would be fairly easy to seal a server behind a sheet of vinyl and purge the air inside with nitrogen. Or just fill the system with dry air if Nitrogen bothers you. The problem now is that the server exchanges air with the larger room, where people are breathing. There is no reason that professionally maintained servers need to be exposed to atmosphere outside of occasional maintenance.

    --
    "Will future ages believe that such stupid bigotry ever existed!" -- Ivanhoe
    1. Re:Nitrogen by ctilsie242 · · Score: 1

      If the server is sunk into the ocean, never to be used again, then it doesn't hurt to purge the air for nitrogen. This stops any corrosion from happening, makes fires impossible, keeps a lot of bacteria types from growing, and lots of other good things.

      There are music studios which fill the mixing room full of nitrogen when it isn't in use, just so the contacts do not corrode.

    2. Re:Nitrogen by Anonymous Coward · · Score: 0

      Seems like they haven't discovered dehumidifiers

    3. Re:Nitrogen by Anonymous Coward · · Score: 0

      Nitrogen is a very effective dehumidifier, and it's cheap. It is commonly used to dehumidify things that are not easily accessed. I've seen it piped into wiring vaults or injected into permeable coax to run the water off that causes impedance changes.

    4. Re:Nitrogen by Pascoea · · Score: 1

      injected into permeable coax

      FM Radio tower feeders are a good example. The cable is massive air-core that is purged/pressurized with nitrogen: Like this stuff

  11. Stupid comment of the day... by Anonymous Coward · · Score: 0

    Making climate change worse... Sink data farms in the ocean and cause further ocean warming.

    1. Re:Stupid comment of the day... by Anonymous Coward · · Score: 0

      Making climate change worse... Sink data farms in the ocean and cause further ocean warming.

      So I assume you don't mind having to burn all that extra fossil fuel to produce the cold air you need to keep the computers running using A/C uints? Just a guess, but It seems to me that this would reduce the CO2 load considerably for a server farm by using a lot less energy.

    2. Re:Stupid comment of the day... by Anonymous Coward · · Score: 0

      Don't respond to his alarmist concerns. He still believes science is magic.

  12. Leave the sea alone? by thePsychologist · · Score: 2

    There are so many lifeforms in the ocean that people rarely see, so if this sinking data center idea takes off, the massive number of sunk data centers could affect these lifeforms and no one will watch out for them.

    Yes, it might reduce the CO2 and barely warm up the sea, but there are other aspects to balancing this equation than just this.

    --
    "What lies behind us, and what lies before us are tiny matters compared to what lies within us." Ralph Waldo Emerson
    1. Re: Leave the sea alone? by Anonymous Coward · · Score: 0

      Data center coral reefs... in the North Atlantic. It could work!

    2. Re:Leave the sea alone? by thegarbz · · Score: 1

      the massive number of sunk data centers could affect these lifeforms

      Indeed. It may give them shelter and a home in an otherwise barren wasteland that is the typical ocean floor.

      Don't get caught in the diving marketing materials. Most of the ocean floor is about as rich in environment as the middle of the Sahara. A completely empty and barren land with nothing but sand, water, and the occasional passing crab. Conversely when humans have in the past sunk things into the ocean they have turned the area in to one teeming with life in the form of what becomes an artificial reef.

      You think housing is bad in Silicon Valley, try being a fish in the ocean.

  13. who? by Anonymous Coward · · Score: 0

    So who's going turn the computers off then on again when the software hangs?

    1. Re:who? by Monkey-Wrench-Inc · · Score: 1

      SpongeBob?

  14. SJW by Anonymous Coward · · Score: 0

    That shows what happens when SJWs take over a technology company!

  15. Project Natick? by GungaDan · · Score: 1

    That's no data center - it's a Twinkie factory!

    --
    Eloi are stupid, throw morlocks at them!
  16. How much energy was consumed by Anonymous Coward · · Score: 0

    Building & setting up this thing versus conventional. Was the energy savings net positive or negative? After all the thing is still consuming the energy to run. Only cooling has been removed. And I thought some of these server farms were already in places like Iceland where all you'd need is to open the windows.

    1. Re:How much energy was consumed by CustomSolvers2 · · Score: 1

      Building & setting up this thing versus conventional. Was the energy savings net positive or negative?

      I was about to write pretty much the same thing. Lots of these "disruptive innovations" only focus on the specific issues which they slightly improve, but not on everything else. Also I find kind of curious that building a container to just lie close to the sea surface is so expensive and demanding. On the other hand, if you have a big research budget to burn, ideally in a publicly noticeable fashion, trying this one doesn't seem a so bad idea.

      --
      Custom Solvers 2.0 = Alvaro Carballo Garcia = varocarbas.
  17. Hope it's for something useful by Anonymous Coward · · Score: 0

    Like cryptomining and warming the ocean for tourists.

  18. What corrosion? by MadMonk67 · · Score: 0

    I've been working in data centers for over 20 years and I've yet to see any corrosion on equipment. If corrosion in your data center is a problem, then you are doing something very wrong. WTF are they talking about?

    1. Re:What corrosion? by Anonymous Coward · · Score: 0

      It's in the sea. Hello?

    2. Re:What corrosion? by Locke2005 · · Score: 1

      If it's hermetically sealed, then corrosion shouldn't be a problem, other than corroding from the outside in, which I'm sure would take longer than the useful life of the equipment sealed inside anyway. However, wouldn't barnacles and other marine life eventually interfere with heat dissipation?

      --
      I've abandoned my search for truth; now I'm just looking for some useful delusions.
    3. Re:What corrosion? by SandorZoo · · Score: 1

      Ben Cutler, who is in charge of Microsoft's Project Natick, is quoted in the article as saying:

      Additionally because there are no people, we can take all the oxygen and most of the water vapour out of the atmosphere which reduces corrosion, which is a significant problem in data centres

    4. Re:What corrosion? by Anonymous Coward · · Score: 0

      It's surprisingly difficult to make something watertight while also having openings for power and data cables. It's the seals for those that are the problem, especially over time.

  19. such gross stupidity by trrosen3908 · · Score: 1

    Apparently MS is unfamiliar with the concept of a boat. Or security.

  20. Why does it need to be deep? by FryingLizard · · Score: 3, Insightful

    Why sink it? That shit sounds expensive. The only thing you're after here is free cooling; why can't it be on the shoreline, or say 50ft offshore? Stick it in a concrete bunker if you like; run a water pump or arrange for natural sea currents to do the work. It's good enough for nuclear power stations.
    This sounds like a toy project.

    --
    [FrLz]
    1. Re:Why does it need to be deep? by b0s0z0ku · · Score: 1

      Yep. Or an attempt to be outside the jurisdiction of any country that has data protection laws.

    2. Re:Why does it need to be deep? by ElizabethGreene · · Score: 3, Insightful

      Or an attempt to be outside the dominion of a country that does /not/ have data protection laws.

      The sword cuts on both sides.

    3. Re:Why does it need to be deep? by WoodstockJeff · · Score: 1

      If it's going to be off-shore, it should be far enough off-shore to be in open ocean, free of territorial waters.

      Sink it far enough below the surface to minimize the effects of storms and to not affect surface navigation, attached to the sea floor by cables.

      It should not affect the wild life, and even could be made big enough to support a crew.

    4. Re:Why does it need to be deep? by Locke2005 · · Score: 1

      Right... because running power and data cables 200 miles offshore is really cheap!

      --
      I've abandoned my search for truth; now I'm just looking for some useful delusions.
    5. Re:Why does it need to be deep? by thegarbz · · Score: 1

      This sounds like a toy project.

      Pilot project. Feasability study. There's plenty of ways to describe such things, but "toy" is not the right one.

      Speaking of toy projects, what was your idea again? Having to purchase ocean front land, build pipework and pumps, more complicated environmental regulations to navigate with NIMBYs to boot.

      Your project just sunk itself (pun intended) unintentionally on cost benefit.

    6. Re:Why does it need to be deep? by AmiMoJo · · Score: 1

      Temperature decreases with depth, down to about 1000m. They must have decided that with the density of servers and amount of heat generated that the temperature on the surface was too high to provide adequate cooling.

      --
      const int one = 65536; (Silvermoon, Texture.cs)
      SJW, n: "Someone I don't like, and by the way I'm a fuckwit" - AC
    7. Re:Why does it need to be deep? by HornWumpus · · Score: 1

      At the Orkneys?

      I'll take a WAG and suggest it's about getting below the weather.

      --
      John McAfee 'It was like that time I hired that Bangkok prostitute; to do my taxes, while I fucked my accountant'
  21. Dead end by Locke2005 · · Score: 1

    Computer equipment depreciates to zero in 2 years. If you can't upgrade the equipment, this is kind of a dead end. Better to locate near geothermal power in Iceland and air cool everything.

    --
    I've abandoned my search for truth; now I'm just looking for some useful delusions.
    1. Re:Dead end by Anonymous Coward · · Score: 1

      The beauty of it is once it's obsolete, it's already been dumped in the sea.

    2. Re:Dead end by HornWumpus · · Score: 1

      You'll get a huge bill from the IRS for trying. 7 years for servers, 3 for desktops/laptops.

      --
      John McAfee 'It was like that time I hired that Bangkok prostitute; to do my taxes, while I fucked my accountant'
  22. "Microsoft sinks data center" by Locke2005 · · Score: 4, Funny

    I'm sure this is by no means the first data center sunk by Microsoft!

    --
    I've abandoned my search for truth; now I'm just looking for some useful delusions.
  23. if you're looking for me by bobmajdakjr · · Score: 1

    you better check under the sea, cause that is where you'll find me underneath the seeaaaaaaaaaalabbbbbbbb underneath the water seaaaaaaaalaaaaaaab at the bottom of, the, sea.

    1. Re:if you're looking for me by Anonymous Coward · · Score: 0

      *INTERNET HIGH FIVE!*

      THERE CAN BE ONLY NONE, DEBBIE! *SMASHES FLASHLIGHT*

      Good lord I loved that show. I still need to go back and watch it again. I think it's up for streaming somewhere.

      captcha: derail

      LOL.

  24. Can’t sweep heat under the rug. by Hallux-F-Sinister · · Score: 1

    Unless 100% of the power used by this kind of data center is renewable, this is like sweeping dirt under the rug. Sure, you’ve dealt with the problem of unsightly dirt on the floor, but at the end of the day the dirt is still there in the room. While dumping the heat from this thing into the vast ocean might seem like a good way to get rid of the heat, the heat remains in the system. Further, layers of ocean water do NOT mix to the point of homogeneity, so the layer where this thing hangs out is going to end up hotter by however much waste heat this thing dumps, with probably far-ranging consequences for marine ecology, changes to ecosystems we depend on to work, the possibility of mass-dyings of populations of plants, animals, fungi, changes in susceptibility to viruses by microbes without which we would all die, etc. etc. etc. according to the Law of Unintended Consequences, which states when you mess with something you can’t possibly know enough about to control all results from your messing, things you have no way of anticipating, often bad, end up happening, and of course the more widely known Murphy’s Law, which states when you sink a data center to save money on discharging heat, something will inevitably go wrong.

    What, you ask? Whatever can. This is potentially going to end up being next-level bad, especially if it seems successful and free of problems, because then EVERYONE will do it, just like dumping all manner of crap into the air, imagining the environment had infinite capacity to cleanse our filth, and we wouldn’t all end up inhaling it and developing cancer, as people have. Also, yes, I know sea != ocean, but waters flow from one to the other in most instances, so the point stands.

    My only hope in all this is that someone does us all the giant favor of coming along and stealing the damned thing, or cutting their cables. Rather than trying to make cooling (and thereby operations cheaper, assuring even more of these heating devices will be made, they should work instead on making them more thermally efficient, denecessitating all the efforts at cooling. Yes, I’m aware I’ve contributed to the problem by using an internet connected device, but not as much as they stand to contribute to the problem by cooking the fish while they’re still IN THE OCEAN.

    --
    Our reign has gone on long enough. Indeed. Summon the meteors.
    1. Re:Can’t sweep heat under the rug. by JustNiz · · Score: 1

      >> so the layer where this thing hangs out is going to end up hotter ...with probably far-ranging consequences for marine ecology,

      Yep agreed. But if undersea volcanic activity is anything to go by, actually more life will live and thrive there.

      > This is potentially going to end up being next-level bad,

      I truly doubt that one small server box could make any real difference to ambient sea temperature more than a handful of feet away, even if it tried. If this experiment starts a trend where companies like Microsoft and Google that have big server farms start moving all their servers underwater, then maybe you'll start to have an actual point.

    2. Re:Can’t sweep heat under the rug. by sloth+jr · · Score: 1

      This is a research project, designed to find answers to those things you called out. In regards to power, MS acknowledges that the majority of power is being consumed by computers, but that a significant portion is dedicated to cooling issues (in a standard DC). Might as well give it a try.

    3. Re:Can’t sweep heat under the rug. by iggymanz · · Score: 2

      hahaha, you know all the power consumed by human civilization, if converted to heat and dumped in the ocean, woul not make an iota of difference compared to the sun. Even the natural variation in solar output totally dwarfs the heat output of mankind. We do NOT have a mans-waste-heat-warming-the-earth problem.

      Pollutions making gases that trap a bit more SOLAR heat, pollution darkening ice and snow to trap more SOLAR heat...yes, those might be a problem.

    4. Re:Can’t sweep heat under the rug. by Anonymous Coward · · Score: 0

      The reality is these datacentres right now are causing far more heat to enter the oceans systems through electricity consumption, manufacturing, disposal etc etc. using natures cooling directly "may" actually be less harmful, though I guess part of these experiments is to measure such things.

    5. Re:Can’t sweep heat under the rug. by Cederic · · Score: 1

      I need to dispel 10 million BTUs. Do I:
      1 - let the sea absorb it
      2 - create 200 million BTUs in wasted frictional energy to provide the cooling needed to dispel it

      Forget the pollution, the greenhouse effects; the efficiency savings alone make this a more sensible idea.

    6. Re:Can’t sweep heat under the rug. by Hallux-F-Sinister · · Score: 1

      >> so the layer where this thing hangs out is going to end up hotter ...with probably far-ranging consequences for marine ecology,

      Yep agreed. But if undersea volcanic activity is anything to go by, actually more life will live and thrive there.

      > This is potentially going to end up being next-level bad,

      I truly doubt that one small server box could make any real difference to ambient sea temperature more than a handful of feet away, even if it tried. If this experiment starts a trend where companies like Microsoft and Google that have big server farms start moving all their servers underwater, then maybe you'll start to have an actual point.

      The difference between this and volcanic activity is important. First, in the case of volcanism, raw materials, nutrients, chemicals, are being added to the system, whereas with the data center, all that's being added is heat. That would be like if the strength of the sun's rays suddenly doubled. Yes, some plants would LOVE that, but many would just get COOKED. Those who can handle the extra light and heat would slowly take over niches previously occupied by those who could not take the heat, but that's kind of the point. What is currently arable farmland would become arid desert, and all the people (and all the animals) who depend on certain plant species being in certain abundance in certain specific locations are simply screwed, and are going to die... or at least adapt to a new reality, VERY rapidly, which often historically has been the death-knell of whole populations of organisms.

      As for the other point, did you read all the way to the end of the sentence you were replying to, where I made the exact point you posited that I would have to be making in order to have a point? Because that's exactly what I said, oddly enough. This becomes next-level bad COMMA... especially if it seems successful and free of problems, because then EVERYONE will do it [...] etc. I don't usually quote myself but... here it seemed appropriate. That had been the point I was trying to convey. One is not a big deal. It is when there are thousands, millions, causing subtle (and later not so subtle) shifts in WHERE in the ocean one may find certain water temperatures, that we run into problems. Fish and other aquatic or marine organisms don't share our ability to live comfortably in places where ambient temps range by nearly a hundred degrees Fahrenheit. We have clothing, fire, air conditioning, paint... and water is something like 40 or 80 (I forget) times more thermally conductive than air, meaning that if you take an entire region of water and jack the temps up a degree or two, (yes, I'm aware that in quintillions of gallons of sea water, that requires immense power,) it ends up forcing the entire population to live elsewhere, either rising a few meters or falling, and any that can't handle that change, and all the attendant other changes that result, are SCREWED .

      Imagine if there were a region of the sea the size of the continental United States, on the sea floor, relatively flat, at a depth of 50 meters. Crawling along the bottom are trillions, maybe quadrillions of individual life forms necessary for the health and ecology of the region, who fix in some way, the chemistry of the water, consuming dead fallen fish, whales, etc., and releasing the sulfur, nitrogen, etc., back into the sea that ends up being needed for all the life that swims above it. Then something happens and the water gets ever so slightly hotter. The temperature there has been nearly constant for lets' say, 700 million years, varying by only a 10th of a degree F. in all that while, and with pressure that is similarly, almost constant. Then suddenly, over a few years or decades, the temp goes up by a half a degree or a degree. The whole sea needn't necessarily increase by that much, due to the effect of layering of water. (I was myself surprised to learn about this kind of thing. He

      --
      Our reign has gone on long enough. Indeed. Summon the meteors.
    7. Re:Can’t sweep heat under the rug. by Hallux-F-Sinister · · Score: 1

      hahaha, you know all the power consumed by human civilization, if converted to heat and dumped in the ocean, woul not make an iota of difference compared to the sun. Even the natural variation in solar output totally dwarfs the heat output of mankind. We do NOT have a mans-waste-heat-warming-the-earth problem.

      Pollutions making gases that trap a bit more SOLAR heat, pollution darkening ice and snow to trap more SOLAR heat...yes, those might be a problem.

      Okay, first, all the power consumeD (past tense) that you're talking about, assuming you're right... but what about 50 years from now, when 95% of human activity is farming cryptocurrency, and there are like, a million computing devices per person...? (Just a guess, I confess, but it seems logical.)

      But again, we can't forget the issue where you can't just assume that after dumping heat in ONE location, it will, completely safely, disperse throughout the system, without impacting all that lives in that place. When a chemical plant dumps its toxic sludge in a river, and that river drains to the ocean, and you say, a few million barrels of MEK (methyl ethyl ketone) in 10^18 gallons (or whatever) of ocean is not going to even be noticeable, not even detectable, but what about the 500 miles of river between where it's dumped and the ocean? And what if it's not MEK but unrefined crude tarsands oil?

      Without (admittedly) looking too far into the details of this plan, I'll bet there's a really good reason they're trying to do this far, FAR away from whatever is LEFT at this point, of the US Environmental Protection Agency, and its ability to regulate shenanigans like this, and fine the fuck out of Microsoft if they turn out to need to be fined, when this ends up going horribly, terribly wrong. Or maybe it's because this experiment might save them money, and they need to do it far away from the IRS's taxing authority, through some chain of subsidiaries and shell companies so they can pretend they didn't profit... but I'm guessing it's more the environmental thing, though I suppose it COULD be both. Or maybe Microsoft just LOOOVVVESSS them some fresh, North Scottish islands fish, and want to go do this where it's freshest? I'm thinking though that it's the lack of rules or taxes thing.

      --
      Our reign has gone on long enough. Indeed. Summon the meteors.
    8. Re:Can’t sweep heat under the rug. by Hallux-F-Sinister · · Score: 1

      I need to dispel 10 million BTUs. Do I: 1 - let the sea absorb it 2 - create 200 million BTUs in wasted frictional energy to provide the cooling needed to dispel it

      Forget the pollution, the greenhouse effects; the efficiency savings alone make this a more sensible idea.

      I need to dispel 10 million BTUs.

      Need to, or want to? I think we should examine this bare assertion a little more closely. Also... do you really think cooling something takes 20 times more power than the thing itself? If you have a thousand watts (1000 W, or 1kW) being used and radiated as heat into the space inside your home from electrical devices running, chugging away, doing their things, and your house is (otherwise) at equilibrium with the outside air, thermally, do you really think it's going to take 20 KILOWATTS to cool your home?!?

      If that's the typical performance you've been seeing, you may need to check your filters, because they may have been surreptitiously replaced with solid blocks of wood... or maybe call the HVAC people, and have the system as a whole checked, because it's not supposed to take that much power to do that. Hell, I'm not sure it should take that much to HEAT the place in the dead of a very cold winter. Perhaps consider having the place insulated? LOL

      Seriously though, my point was mainly that they're monkeying with things they don't understand, in the name of corporate profits, and when that's what you're concerned about, corporate profits, anything that gets in the way that you can pretend doesn't exist, is going to be treated like... well, like it doesn't exist. Sadly, things don't just cease to exist because they're inconvenient.

      --
      Our reign has gone on long enough. Indeed. Summon the meteors.
    9. Re:Can’t sweep heat under the rug. by Cederic · · Score: 1

      My point is that your point is irrelevant because you can not cool something below ambient (let alone quickly) without suffering inefficiencies. So you will always use more energy to cool something than it will dissipate if it cools naturally.

      So the ocean may warm up but less than if it was cooling the power station used to drive aircon.

      20x? Depends how much of the power station, fuel mining, transportation, transmission infrastructure (and associated mining, transportation, manufacturing), power conversion, aircon efficiency (and manufacture, Inc. mining, trans.. etc) and severe other uses of energy that you want to allocate as part of calculating it.

      I haven't run the maths. I don't need to. My point stands anyway, and proves the flaw in your own.

    10. Re:Can’t sweep heat under the rug. by Hallux-F-Sinister · · Score: 1

      My point is that your point is irrelevant because you can not cool something below ambient (let alone quickly) without suffering inefficiencies. So you will always use more energy to cool something than it will dissipate if it cools naturally.

      So the ocean may warm up but less than if it was cooling the power station used to drive aircon.

      20x? Depends how much of the power station, fuel mining, transportation, transmission infrastructure (and associated mining, transportation, manufacturing), power conversion, aircon efficiency (and manufacture, Inc. mining, trans.. etc) and severe other uses of energy that you want to allocate as part of calculating it.

      I haven't run the maths. I don't need to. My point stands anyway, and proves the flaw in your own.

      No it doesn't, especially when you haven't done the math. Suggesting or implying I have to do your math for you undermines your argument. Come back with math complete and we'll talk. P.S., why do you think using seawater magically undoes the reality of having to expend energy to cool something to below ambient? Why would you imagine seawater has some magical property of heat elimination that air does not? Sure it's denser, and thermally more efficient, but as your "datacenter" heats the water around it, the heat doesn't just magically disperse instantly, and as the water around it warms, this will reduce the efficiency with which this thing can dump its heat, requiring either that the target for how cool it's kept has to be adjusted (up) or that it be built with a really BIG heatsink to increase the contact between it and the water. Oh, and what about the cost of mining all that metal you need for that heatsink? Or are you planning on building a fan into yours, or a propeller, or an impeller, and forcing water through it, in which case, where does the energy come from to do that?

      Simply asserting that you've "won" without presenting evidence in an argument does not make the point of someone you disagree with go away. But I'm done trying to educate you. My argument doesn't have a "flaw" which you've "proven". My argument (perhaps maybe you've forgotten) was that relocating the thing that is dumping heat to be closer to a place that can absorb large amounts of it will turn out to be bad for the environment and people who depend on it, especially if it turns out to be technically feasible and economically viable, because everyone will start doing it, and there will be tons of these damned things, everywhere, concentrating heat in places where that heat will cause changes I'm confident, without breaking into their offices and looking, they haven't fully investigated because if they HAD, they might have learned something they will get sued for having known BEFORE they did it.

      Good (for-profit) corporations are great at not finding out what they suspect their executives may not be able to buy their way out of trouble for knowing at the time down the line.

      What they're planning to do is likely hazardous to the things that live where they're planning to do it, it's being done in a place no human knows enough about to do this safely, as the seas and oceans are inadequately explored or understood, they're vast systems existing in delicate balances in probably a hundred different ways, but it will help their bottom line as a corporation, so they don't care about little things like, what they're doing potentially devastating others, because they live for the here and now. "Damn the torpedoes and full speed ahead" only means what it does because the moron who said it got lucky and most of his boats didn't sink that particular day. If they had, the saying would have the more appropriate meaning, being an expression of hubris that gets other people hurt or killed because of disregarding the consequences.

      Perhaps the takeaway from this should not be people arguing about the technical points when neither knows anything about them, but that a way needs to be found to make enviro

      --
      Our reign has gone on long enough. Indeed. Summon the meteors.
    11. Re:Can’t sweep heat under the rug. by Cederic · · Score: 1

      So to recap. You're pissed off that someone is doing something that will improve the environmental friendliness of providing data centre services because it will also be economically beneficial for them and they might make higher profit.

      I'll just highlight two reasons you're talking shit:
      1 - by suggesting this will be more profitable for them you've de facto acknowledged that it's more efficient due to the lower costs that offers
      2 - they're doing tthis in tidal waters which provides substantially more water flow than a whole fucking barrage of water pumps

      So even if they know nothing about the sea (demonstrably bollocks) there's a net benefit to the environment. Running the numbers merely tells us the magnitude of that benefit; raw fucking logic tells us it exists.

      Try some, you might find you like it.

    12. Re:Can’t sweep heat under the rug. by JustNiz · · Score: 1

      I think you're massively overreacting and doomsaying. Also... all your arguments are based on conjecture and suppositions...
      In fact there are many natural sources of heat and cooling affecting ocean temps already much more and on a much larger scale than a few server boxes could.

  25. Wtf by xettera · · Score: 1

    ..because we're not warming the oceans fast enough already?

  26. ZERO security risk... by ElitistWhiner · · Score: 1

    Well-Cache ( i.e. water well, ocean depth) - hermetically sealed hi-value water storage scheme in which any breach by design self-destructs its contents that provides a Zero risk highly secure environment. ZERO risk based on the NIST Common Misuse Scoring System (NISTIR 7864)

  27. whoa... by Anonymous Coward · · Score: 1

    I guess sinking Github wasn't enough for 'em...

  28. Headline too long by mi · · Score: 1

    waterboarding?

    That's true about all computer-cooling, though...

    Personally, I think, the headline should just read "Microsoft Sinks Data Centre".

    --
    In Soviet Washington the swamp drains you.
  29. Let the puns begin... by MrKaos · · Score: 1

    Cool.

    --
    My ism, it's full of beliefs.
  30. Github data center by Anonymous Coward · · Score: 0

    It was the data center hosting Github. They wanted to clean out all of the open source cancer.

  31. Warming oceans? by Anonymous Coward · · Score: 0

    Does the idea of warming oceans and upsetting water currents with massive undersea data centers sound bad to anyone... yet?

    1. Re:Warming oceans? by HornWumpus · · Score: 1

      No.

      Does the idea of learning thermal physics sound bad to you?

      --
      John McAfee 'It was like that time I hired that Bangkok prostitute; to do my taxes, while I fucked my accountant'
  32. Windows source code by Anonymous Coward · · Score: 0

    ...but if the computers onboard break, they cannot be repaired.

    If Microsoft could somehow put every copy of the Windows source code on this, then I'm all for it.

  33. Things grow, where ever. by Anonymous Coward · · Score: 0

    Living in a coastal town and seeing what happens whenever there is a thermal or nutritional gradient, I will watch this with a sense of humor.

  34. HFT arbitrage via midpoint of ocean cable by Anonymous Coward · · Score: 0

    If you can execute HFT arbitrage from a mid ocean sunken datacenter sitting on a fiber optic sea cable, the physics of transmission means you will win over HFT firms that have to use the whole cable.

    Say if you are working high frequency trading arbitrage between the Chicago Mercantile Exchange and London, sitting in the middle of the Atlantic means you can beat US domestic traders trying to work with London, or London traders trying to work Chicago.

    There are other latency sources you need to account for (fiber optic repeaters, other networking gear) but raw physics says the closer trader wins the arbitrage trades. The only way to beat that would be either via shortwave radio bouncing off the ionosphere (less relay) or neutrino transmission assuming that is less than the actual achieved latency from mid-atlantic to a bourse.