• Microsoft ends underwater “Cloud” server-farm trial with promising results.

    Home » Forums » Outside the box » Rumors and what-ifs » Microsoft ends underwater “Cloud” server-farm trial with promising results.

    Author
    Topic
    #2296768

    The experiment carried out by MS offshore the Orkney islands with a submarine server farm cooled by the water around it and powered by locally produced renewable energy from solar and wind has been concluded with promising results, according to this report:

    https://www.bbc.com/news/technology-54146718

    Excerpt:

    At MS “Their first conclusion is that the cylinder packed with servers had a lower failure rate than a conventional data centre.
    When the container was hauled off the seabed around half a mile offshore after being placed there in May 2018, just eight out of the 855 servers on board had failed.
    That compares very well with a conventional data centre.
    “Our failure rate in the water is one-eighth of what we see on land,” says Ben Cutler, who has led what Microsoft calls Project Natick.
    The team is speculating that the greater reliability may be connected to the fact that there were no humans on board, and that nitrogen rather than oxygen was pumped into the capsule.
    “We think it has to do with this nitrogen atmosphere that reduces corrosion and is cool, and people not banging things around,” Mr Cutler says.

    Ex-Windows user (Win. 98, XP, 7); since mid-2017 using also macOS. Presently on Monterey 12.15 & sometimes running also Linux (Mint).

    MacBook Pro circa mid-2015, 15" display, with 16GB 1600 GHz DDR3 RAM, 1 TB SSD, a Haswell architecture Intel CPU with 4 Cores and 8 Threads model i7-4870HQ @ 2.50GHz.
    Intel Iris Pro GPU with Built-in Bus, VRAM 1.5 GB, Display 2880 x 1800 Retina, 24-Bit color.
    macOS Monterey; browsers: Waterfox "Current", Vivaldi and (now and then) Chrome; security apps. Intego AV

    1 user thanked author for this post.
    Viewing 5 reply threads
    Author
    Replies
    • #2296777

      No-one considered how it warmed up the Pentland Firth.

      Dell E5570 Latitude, Intel Core i5 6440@2.60 GHz, 8.00 GB - Win 10 Pro

      2 users thanked author for this post.
      • #2296779

        No mention of an environmental impact study, either.

        Ex-Windows user (Win. 98, XP, 7); since mid-2017 using also macOS. Presently on Monterey 12.15 & sometimes running also Linux (Mint).

        MacBook Pro circa mid-2015, 15" display, with 16GB 1600 GHz DDR3 RAM, 1 TB SSD, a Haswell architecture Intel CPU with 4 Cores and 8 Threads model i7-4870HQ @ 2.50GHz.
        Intel Iris Pro GPU with Built-in Bus, VRAM 1.5 GB, Display 2880 x 1800 Retina, 24-Bit color.
        macOS Monterey; browsers: Waterfox "Current", Vivaldi and (now and then) Chrome; security apps. Intego AV

        1 user thanked author for this post.
      • #2296968

        (humour alert) Aaah, so thats the reason why ice on the north pole is melting so fast 🙂

        Dell Latitude 3420, Intel Core i7 @ 2.8 GHz, 16GB RAM, W10 22H2 Enterprise

        HAL3000, AMD Athlon 200GE @ 3,4 GHz, 8GB RAM, Fedora 29

        PRUSA i3 MK3S+

    • #2296851

      just eight out of the 855 servers on board had failed.
      That compares very well with a conventional data centre.
      “Our failure rate in the water is one-eighth of what we see on land,” sa

      that seem s like they expect a very high failure rate~~

      🍻

      Just because you don't know where you are going doesn't mean any road will get you there.
      1 user thanked author for this post.
      • #2296900

        Wavy: “Oscar CP wrote: just eight out of the 855 servers on board had failed

        (Disclaimer: I’m given too much credit here, because did not write, merely quoted, without the added emphasis). As to the ultimate significance of those server failures, I do believe that these can be made less important by keeping a direct proportion, by design, between their number and the numbers written on customer’s bills.

        On the related issue brought up by Rick Corbett of what they learned at MS after two years of running this experiment: My guess is that, first of all, at MS they are just happy they got their underwater equipment back in one piece, because it did not blow up, developed water leaks that destroyed its high-tech contents, or was scrunched down by a giant octopus. Learning that was to make progress. Also learned was that they could run all those servers with, as reported, an unusually low number of problems while using electrical power from only renewable “green” sources dependent on the variable availability of sunlight and, or wind. (Although I believe that wind is not often in short supply in the Orkneys.)

        Also learned was the useful fact that filling a server farm with nitrogen rather than air and not having many people working in it is probably very good for the equipment. Although breathing nitrogen might not be quite so wonderful for the employees. But I would imagine that could be solved by writing their contracts with the appropriate disclaimers and company immunity clauses.

        Finally, it is my understanding that the fish were not consulted and kept their opinions to themselves. Well, that’s fish for you.

        Ex-Windows user (Win. 98, XP, 7); since mid-2017 using also macOS. Presently on Monterey 12.15 & sometimes running also Linux (Mint).

        MacBook Pro circa mid-2015, 15" display, with 16GB 1600 GHz DDR3 RAM, 1 TB SSD, a Haswell architecture Intel CPU with 4 Cores and 8 Threads model i7-4870HQ @ 2.50GHz.
        Intel Iris Pro GPU with Built-in Bus, VRAM 1.5 GB, Display 2880 x 1800 Retina, 24-Bit color.
        macOS Monterey; browsers: Waterfox "Current", Vivaldi and (now and then) Chrome; security apps. Intego AV

    • #2296852

      So, basically, Microsoft has no clue after ~2 years of study…?

    • #2296929

      Also learned was the useful fact that filling a server farm with nitrogen rather than air and not having many people working in it is probably very good for the equipment

      Nope, we learned that one example performed well but we don’t know if it was the people, the nitrogen or the temperature.

      A sample size of one does not a study make.

       

      cheers, Paul

      • #2296941

        One study, not one data point, and one that must be the beginning of a series, if this is a serious project,

        And, please, notice that I write “probably”, not “definitely” or “unquestionably”, because what I think might have been learned so far (“might” because I was not there), besides the fact that the thing did not get destroyed for unexpected reasons that could become obvious only afterwards, is that the idea of putting the server farm in a watertight containing vessel on the sea floor so it gets cooled by the surrounding ocean acting as a huge heat sink seems to have worked as planned, that using power from renewable sources in  the underwater installation created no serious problems in some two years of (as far as I know) continuous functioning, so it is probably a practical idea to be tested further, that using nitrogen to avoid electrical fires seem to have worked without undesirable side effects, so it is probably another good one, and that not having people inside probably helped reduce the number of failures. If all this is true, then it gives a direction for future research, with those four factors chosen to be studied further and perhaps with a larger experimental installation, while de-emphasizing other things that do not seem to have been as relevant. And most likely there is a great deal of telemetry data still to be analyzed as well, some of it to be used to model and redesign parts of the equipment with the help of computer simulations. All of which might or might not point the right way to go, but that is how research often has to be done in engineering when trying out some new ideas in the real world, with all its complexities, particularly at the beginning. The process can be long and expensive, with success not guaranteed.

        Ex-Windows user (Win. 98, XP, 7); since mid-2017 using also macOS. Presently on Monterey 12.15 & sometimes running also Linux (Mint).

        MacBook Pro circa mid-2015, 15" display, with 16GB 1600 GHz DDR3 RAM, 1 TB SSD, a Haswell architecture Intel CPU with 4 Cores and 8 Threads model i7-4870HQ @ 2.50GHz.
        Intel Iris Pro GPU with Built-in Bus, VRAM 1.5 GB, Display 2880 x 1800 Retina, 24-Bit color.
        macOS Monterey; browsers: Waterfox "Current", Vivaldi and (now and then) Chrome; security apps. Intego AV

    • #2296974

      (humour alert) Aaah, so thats the reason why ice on the north pole is melting so fast 🙂

      You are not far from the troth

      Wonder how much marine life was killed due to heat. pollution, radiation…

      https://www.nbcnews.com/science/environment/dismay-huge-chunk-greenlands-ice-cap-breaks-rcna117

      • This reply was modified 4 years, 7 months ago by Alex5723.
      • #2297075

        If the terse comment by Alex ( #2296974 ) is on the MS project near the Orkney islands, then there was no need for having a radiation source in the sunken installation and no mention of one anywhere I’ve looked, as this topic interests me as an engineer,  while probably no one knows if heating from the underwater server farm was an issue or not for marine life, because the environmental impact, as far as I am aware at least, may not have been considered or, if considered and also carried out, the results have not yet been made known to the public.

        If it is not about this MS project, then are radiation and global warming issues affecting not just the polar regions but the world at large? Yes, plain common sense and much hard evidence, respectively, indicate that they are, indeed: from slowly rotting away sunken nuclear submarines with nuclear fuel on board and in any nuclear weapons they were carrying when they sunk, to the effect of higher water temperatures intensifying storms and that of the excess CO2 dissolved in the oceans causing their acidification, lethal to many organisms with calcium-based shells and corals, as well as rising sea levels. But this is not the topic I had in mind when I started this thread.

        Ex-Windows user (Win. 98, XP, 7); since mid-2017 using also macOS. Presently on Monterey 12.15 & sometimes running also Linux (Mint).

        MacBook Pro circa mid-2015, 15" display, with 16GB 1600 GHz DDR3 RAM, 1 TB SSD, a Haswell architecture Intel CPU with 4 Cores and 8 Threads model i7-4870HQ @ 2.50GHz.
        Intel Iris Pro GPU with Built-in Bus, VRAM 1.5 GB, Display 2880 x 1800 Retina, 24-Bit color.
        macOS Monterey; browsers: Waterfox "Current", Vivaldi and (now and then) Chrome; security apps. Intego AV

    • #2297077

      Perhaps a desalination plant in conjuction with the server farm would be beneficial? Hydro/Wave/Solar/Wind energy providing the electrical supply with fresh water as a product to reduce the rising sea levels due to climate change 😛
      I’m sure I read somewhere that Iceland is now a growing location for server farms..hmmm

      Windows - commercial by definition and now function...
      1 user thanked author for this post.
    Viewing 5 reply threads
    Reply To: Microsoft ends underwater “Cloud” server-farm trial with promising results.

    You can use BBCodes to format your content.
    Your account can't use all available BBCodes, they will be stripped before saving.

    Your information: