Stories
Slash Boxes
Comments

SoylentNews is people

posted by hubie on Tuesday November 22, @10:53PM   Printer-friendly
from the server-crash-and-burn-means-something-different-now dept.

https://www.extremetech.com/extreme/341033-eu-mulls-putting-data-centers-in-space

The world's ever-increasing reliance on the internet comes at a physical cost. Data centers, which fulfill the vital role of housing and maintaining core computer services and data, are a central element of any operation that relies on digital resources. They're also physically large; as an organization grows, it does, too. Eventually, organizations are forced to consider not only where to put their data centers, but also how to power them efficiently and how to mitigate their emissions.

In recent years, we've started to stick data centers in deserts or in the middle of the ocean. Deserts present few service-disrupting natural disasters and tend to provide plenty of solar power; the ocean, as with Microsoft's Project Natick, helps keep data centers cool. But desert data centers are still land-intensive, and no Earth-based data centers are without their emissions...the key word, of course, being "Earth-based." The European Union thinks it can beat this challenge by sending data centers into space, and it's already working on testing this theory through a $2 million study called ASCEND.

Advanced Space Cloud for European Net zero emission and Data sovereignty (ASCEND) is the brainchild of the EU and Thales Alenia Space, a European aerospace company. The study's goal is to explore the feasibility of placing data center stations in low Earth orbit (LEO). [...]

If ASCEND is successful, the resulting technology could contribute to Europe's goal of achieving carbon neutrality by 2050 under the Green Deal.


Original Submission

This discussion was created by hubie (1068) for logged-in users only. Log in and try again!
Display Options Threshold/Breakthrough Mark All as Read Mark All as Unread
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
(1)
  • (Score: 2, Troll) by Anonymous Coward on Tuesday November 22, @11:22PM

    by Anonymous Coward on Tuesday November 22, @11:22PM (#1281172)

    Grab all the politicos in Brussels, and shoot them into space. Data centers are far less toxic than politicians!

  • (Score: 2) by coolgopher on Tuesday November 22, @11:29PM (5 children)

    by coolgopher (1157) Subscriber Badge on Tuesday November 22, @11:29PM (#1281173)

    Can I sign up to be a field technician for these?

    • (Score: 3, Funny) by captain normal on Wednesday November 23, @01:30AM (2 children)

      by captain normal (2205) on Wednesday November 23, @01:30AM (#1281185)

      The commute would be terrible.

      --
      “I have not failed. I’ve just found 10,000 ways that won’t work.” Thomas Edison
      • (Score: 2) by looorg on Wednesday November 23, @02:00AM

        by looorg (578) on Wednesday November 23, @02:00AM (#1281189)

        but think of all the miles you'll be getting on your cc.

      • (Score: 2) by DannyB on Wednesday November 23, @03:24PM

        by DannyB (5839) Subscriber Badge on Wednesday November 23, @03:24PM (#1281286) Journal

        The ping times would be more terrible than commute times. At least during the commute there would be space toilets or diapers.

        The technicians adjusting the knobs to control the ping times might accidentally on purpose turn the knob too far and get slightly negative ping times. How long this would go undetected would be a mystery. Whether or not anyone would need to modify ping implementations to correctly report timings under these conditions is unknown.

        --
        I get constant rejection even though the compiler is supposed to accept constants.
    • (Score: 3, Funny) by c0lo on Wednesday November 23, @08:54AM (1 child)

      by c0lo (156) on Wednesday November 23, @08:54AM (#1281244) Journal

      Are you the xkcd705 [xkcd.com] type?
      Because we don't have budget for shoes, we mostly spent it on the study.

      --
      https://www.youtube.com/watch?v=aoFiw2jMy-0
  • (Score: 5, Insightful) by higuita on Wednesday November 23, @12:01AM (1 child)

    by higuita (2465) on Wednesday November 23, @12:01AM (#1281176)

    this can't be real, this is just a study for some "what if", specially if there is new tech that helps solving this problems:

    - sending things to space cost a lot, for each Kg you have to push, you will need more Kg of fuel
    - that alone makes everything not ecological, as that fuel is produced mostly in non-renewable ways
    - as it cost a lot, you can't send big things, so in best case, a small data center... and that takes no space in earth, a car garage is enough to replace a "space data center"
    - maintenance... satellites and likes have multiple redundancies so if something fails, other take over... for a "data center", this is even worse, most hardware is build so human can act on failure, in space that needs to be automatic, so special hardware needs to be build, putting again the cost up
    - even low orbit, the "data center" would be log more exposed to radiation and so failure and data corruption

    So i bet this was just one excuse for someone to get a researcher grant

    • (Score: 2) by c0lo on Wednesday November 23, @09:09AM

      by c0lo (156) on Wednesday November 23, @09:09AM (#1281246) Journal

      - sending things to space cost a lot, for each Kg you have to push, you will need more Kg of fuel

      Ummm.... ok, more, but more than what?
      Because iff you can reach the point in which your more is actually less that the amount of energy consumed by that tech-kg on Earth, placing that kg in orbit is an one-off investment with a positive return.

      Hint: I was looking for an arrgument along the line of "if you have a tech with such a low energy consumption it can work on the puny solar panels you can afford to launch in space, it is likely you can afford to use it with the solar panels on Earth surface too. Launching an SSD in space doesn't magically make it more energy efficient", I found none.

      --
      https://www.youtube.com/watch?v=aoFiw2jMy-0
  • (Score: 4, Insightful) by thatsme on Wednesday November 23, @01:21AM (1 child)

    by thatsme (6857) on Wednesday November 23, @01:21AM (#1281183)

    $2 million study?

    Buy any engineer a few beers and he'll do the math and show that this makes no sense whatsoever.

    • (Score: 0) by Anonymous Coward on Wednesday November 23, @08:37PM

      by Anonymous Coward on Wednesday November 23, @08:37PM (#1281332)

      Buy any engineer a few beers and he'll do the math and show that this makes no sense whatsoever

      Ah, but that isn't the point of the exercise though, is it?

      As close on 30 years have passed, I think I can safely admit under the cloak of ACdom that I once had a hand in writing up a bid for some of the lovely €U pork that was up for grabs as part of a frankly insane sounding feasibility study for the computerisation and centralisation of a traditionally low tech and localised industry on a pan European basis, but still allowing for the maintenance of a high degree of local and regional autonomy.

      We thought our bid stood a better chance than most, partially as it was a joint venture between three EU member states and they used to just love that sort of shit, but mostly because one of our main participants then girlfriend just happened to be the daughter of his country's head honcho.

      Turns out the whole thing was a load of cobblers and we'd no chance, word came down via a channel mentioned above that the fix was in as the scheme was dreamt up by some Brussels apparatchik as a means to funnel €U pork to friends of his in that industry back in the old country whilst also creating more work for said apparatchik's fiefdom, thus furthering his way up the Brussels bureaucratic greasy pole. No sour grapes involved here, after all, we were playing the same game but unfortunately for us we didn't know how stacked the deck was, the apparatchik and his connections trumped our head honcho's daughter card.

      The monies involved were not insubstantial, and, needless to say, said study went the way of the fabled ouzelum bird, so much so I can't find hilt nor hair of it online.

  • (Score: 2) by progo on Wednesday November 23, @02:50AM (2 children)

    by progo (6356) on Wednesday November 23, @02:50AM (#1281191) Homepage

    I can't get past the headline.

    • (Score: 2, Informative) by Anonymous Coward on Wednesday November 23, @03:14AM

      by Anonymous Coward on Wednesday November 23, @03:14AM (#1281195)
      On the contrary, radiating is just about the only method you have for getting rid of waste heat when you're in space, and that is an incredibly inefficient way of doing it.
    • (Score: 3, Interesting) by ElizabethGreene on Thursday November 24, @12:41AM

      by ElizabethGreene (6748) on Thursday November 24, @12:41AM (#1281363)

      You've hit the nail on the head. Thermal management in space is extremely hard. If you google/bing a photo of ISS you can see the size of its heat radiators. They are the long wiggle-folded flat structures. IIRC they are 40 feet long and about 4 feet wide. The total heat rejection capacity of the external active thermal control system (EATCS) is about 70KW. That's comparable to a semi-truck radiator here in the atmosphere.

  • (Score: 0) by Anonymous Coward on Wednesday November 23, @03:12AM (2 children)

    by Anonymous Coward on Wednesday November 23, @03:12AM (#1281194)

    Space is hell for electronics, with hard radiation everywhere and the only way you can dissipate excess heat is by radiative transfer (no atmosphere or anything else to dissipate heat out). There's a reason why even a recent space probe like the ESA's Solar Orbiter (launched in 2020) still uses a CPU that is more or less the equivalent of a SPARC from a Sun workstation of the early 1990s. If your electronics aren't hardened, you'll need to shield them, but that increases the weight of your spacecraft, and getting something heavy into space generally means the expenditure of still more non-renewable, non-carbon neutral energy. Also, what method do you have for transmitting data to and from your orbital data centre that does not have the high latency that satellite communications tend to have?

    I don't see why you need a $2 million study to figure out that this idea is a total non-starter.

    • (Score: 2) by takyon on Wednesday November 23, @04:58AM (1 child)

      by takyon (881) <{takyon} {at} {soylentnews.org}> on Wednesday November 23, @04:58AM (#1281206) Journal

      Latency is not such a big deal in low Earth orbit. It might even be superior to intercontinental undersea cables, see discussion surrounding Starlink. Radiation might be manageable, and future innovations like nanoscale vacuum-channel transistors could help more.

      It's the heat dissipation that makes it a non-starter. There are many things [microsoft.com] you can do on Earth to manage it. The satellite would have to add more mass in the form of giant radiators, or run cooler (lower performance).

      --
      [SIG] 10/28/2017: Soylent Upgrade v14 [soylentnews.org]
      • (Score: 2) by ElizabethGreene on Thursday November 24, @12:45AM

        by ElizabethGreene (6748) on Thursday November 24, @12:45AM (#1281364)

        The opposite direction is a possible solution too. Thermal Radiation scales with the fourth power of temperature, so hotter servers (or more realistically clever hardware design that allows a very high working fluid temperature) could help.

  • (Score: 0) by Anonymous Coward on Wednesday November 23, @03:31AM

    by Anonymous Coward on Wednesday November 23, @03:31AM (#1281197)

    Pre-WWII this fighter plane was built, https://en.wikipedia.org/wiki/Curtiss-Wright_XP-55_Ascender [wikipedia.org]
    The prop is at the rear, it's a "pusher" aircraft.

    Insiders quickly started to call it the "ass-ender".

    We can take a cue from history and call this satellite the Ass-end.

  • (Score: 2) by istartedi on Wednesday November 23, @04:52PM

    by istartedi (123) on Wednesday November 23, @04:52PM (#1281290) Journal

    One of the unintended consequences of making environmental policy almost exclusively based on CO2 emissions is that such policies are often worded in such a way as to create accounting shuffles that trade other forms of environmental harm for reduced CO2 output, or perhaps actually increase CO2 output by running it somewhere else off the books.

    I defy you to tell me that a space-shot data center puts out less CO2 than an earth-based solar powered data center. "Oh look, no CO2 from ongoing operations". Yeah. The launch? The satellite up/down link? Bite me. It's almost certainly a scam.

    It's not the worst though. The one that always gets me is the one where they cut down trees, shipped them across an ocean and burned them because "firewood is carbon neutral".

    F*** me. So many scams.

(1)