Forgot your password?
typodupeerror
IBM Hardware

IBM Deploys Hot-Water Cooled Supercomputer 112

Posted by samzenpus
from the hot-is-the-new-cool dept.
MrSeb writes "With the ISC (International Supercomputer Conference) kicking off this week, there's been a flurry of announcements around new supercomputer buildouts. One of the more interesting systems debuting this week is SuperMUC — IBM's new supercomputer at the Leibniz Supercomputing Center in Germany IBM is billing SuperMUC as the first 'hot-water cooled supercomputer,' an advance it claims cut power consumption by 40%. Dubbed Aquasar, the new system looks like any standard water cooler: water is pumped in one side of the blade, circulates throughout the system, and is pumped out. The difference, according to IBM, are the microchannels etched into the copper heatblock above the CPU cores. Rather than simply being dumped, SuperMUC's waste heat is designed to be converted into building heat during winter. Presumably it is mostly radiated away in summer, rather than being dumped into the offices of angry German scientists."
This discussion has been archived. No new comments can be posted.

IBM Deploys Hot-Water Cooled Supercomputer

Comments Filter:
  • What again? Someone call Chet Heath!
    • by bmo (77928)

      CURSE YOU!

      *shakes tiny fist*

      --
      BMO

      • by Taco Cowboy (5327) on Monday June 18, 2012 @08:47PM (#40365435) Journal

        Laundry shops need a lot of very hot water to "cook" the dirty linen that they receive everyday

        Almost all laundry shops are using water heaters - whether they be electric powered water heaters or gas-burner powered water heaters - to heat up the water

        Here's my proposal:

        On the side of all super-computer center or any large scale data-center the authority should draw up a special "zone" for laundry shops

        That way, all the hot water generated from the computers will be put to good use - without any additional wastage of precious energy resource

        It's kinda fits into the "Go Green" concept that is so popular these days
         

        • by bmo (77928)

          Japanese baths.

          Great, big, Japanese baths.

          Think about it!

          --
          BMO

          • by Taco Cowboy (5327)

            There's one big problem

            If you get a Japanese Bath (with beautiful Japanese ladies enjoying their baths) on the side of a super-computer center or data-center, who will man the data-center?

            All the geeks from the data-center will flock to join the Japanese ladies in the Japanese Bath

            • by Anonymous Coward

              Not to mention all the additives that will go into the water, unless you are planning to have a separate water supply and system that transfers the heat between them.

              The perk for using it as a heat source for a HVAC system is that the circuit doesn't need to be open, unless you want to breath fluid. My college has a system like this for their server rooms and all the floors. If memory serves me right, they supply heating to only half of the building at a time (north or south), because of the way the sun hit

            • All the geeks from the data-center will flock to join the Japanese ladies in the Japanese Bath

              And even with the ladies already naked I fear the geeks still got no shot....
              How about building REALLY SMART Hot Dog carts? With that much hot water we could have ever conceivable version of tube steak hot and ready to go.

              • All the geeks from the data-center will flock to join the Japanese ladies in the Japanese Bath

                And even with the ladies already naked I fear the geeks still got no shot.... How about building REALLY SMART Hot Dog carts? With that much hot water we could have ever conceivable version of tube steak hot and ready to go.

                Hot dog carts were banned in my town as a result of the G-string clad cart employees causing many automobile accidents.

        • by anubi (640541)
          Excellent insight there.

          I am in the process of redoing my A/C so that I use a liquid-to-liquid heat exchanger ( tube-in-tube; actually mechanized as a length of copper tubing in a rubber hose ) so I can take advantage of the approx 60 deg F city water.

          I figure if I am going to irrigate, the plants won't mind it at all if I warm it to about 85-90F. The refrigeration loop runs on propane - plain old barbeque gas no less. As a refrigerant, not a fuel.

          The main concern I have had is the refrigerant loop
          • by Taco Cowboy (5327)

            beaming the heat to deep space

            I don't think solar panel can beam heat into deep space

            • by anubi (640541) on Tuesday June 19, 2012 @01:35AM (#40366819) Journal
              At night, my roof faces the black sky.

              I notice my car gets wet from condensation. Its "beaming" its thermal energy off into the night sky - just as it will accept energy from the sun during the day, becoming quite hot.

              I figure if I am streaming 90 degree water into a solar collector at night, it may cool it off to 80 degrees or so - especially if I combine it with evaporative cooling.

              Its the typical "it gets cold at night" thing.

              If you were in outer space with an infrared detector over my house at night, you should see my solar collector "lit up".

              The idea is I have a lot of BTU I want to get rid of in my quest to liquefy propane gas at high pressure. I can heat up air ( conventional method of doing it ), or transfer the heat to a cooler mass, ( water in my case ), evaporate water with it, and I want to experiment to see how much BTU I can radiate with a standard solar collector panel ( the ones with liquid channels ).

              I would like to experiment with standard PV panels bonded onto liquid-channel panels so that during the day, the PV makes electricity, while the liquid panels not only cool the PV array, but provide preheat for a 100 gallon water tank.

              Fluid circulation pumps will route water from the tank, through the collector, then back to the tank as long as collector temperature exceeds tank temperature.

              Of course, once the sun sets, the panel is no longer experiences an influx of about 1KW/m^2 solar energy.

              At night, it will cool off and become quite cold all by itself as it faces the night sky. That's when I am going to attempt to heat the panel back up by circulating water used to cool the propane exchanger ( condenser ). I see it as about 100 square foot of blackbody radiator. What I want is some practical experience on how many BTU I can get rid of doing it this way, as the circulator pumps draw much less energy than the fans required to move the air in a liquid-air exchanger.

              I already have an aluminum roof. That thing gets so cold during the summer I have been having a problem with condensation causing mildew problems. Despite outside air temps of 80-90 F. For condensation to form, the roof has to drop below dew point temperature, and judging from how fast I am condensing liquid water from the air, I get a gut feeling I am already beaming out quite a lot of heat.

              So, in a sense, I am "beaming" the energy to deep space just as a light bulb "beams" the energy of its heated filament into a dark room.

              By far, the most practical is to simply evaporative cool the system... but what if water is not freely available ( design for the Middle East. ).

              That is what I liked about your post. You saw the heat being generated in a server farm, and noted it was just the right temperature for use in a laundromat. A helluva lot of BTU that could have been used - wasted. If more people had your mindset, we could enjoy our creature comforts without paying twice for energy. It simply doesn't make sense to waste things.
          • by Teun (17872)
            You have some good tech but strange ideas re. saving on energy.

            Getting rid of excessive heat is to be done by reusing it, not by loosing it into space (or more likely the atmosphere).
            You write about the oil price, the installation of original article is positioned in Germany, one of the many countries that is seriously worried about an irrevocable human effect on the climate also know as the green house effect.
            So Germany has like all other EU countries decided to promote renewable energy and best practis

          • by Muad'Dave (255648)

            ... use solar collectors in reverse at night to get rid of heat...

            I'm fairly certain you can. Do you have one of those contactless IR thermometers? Point it at a clear blue sky - it'll read way below zero (F or C). You're reading the effective temp of the atmosphere up to space, and the efficiency of your radiator is proportional to the temperature difference. If you point it at the base of an overhead cloud, it reads much higher.

        • by mspohr (589790)

          All hotels have air conditioning equipment which generates waste heat.
          They also have laundries and guest hot water which need a lot of heat.
          How stupid not to combine them?

          • by Taco Cowboy (5327)

            All hotels have air conditioning equipment which generates waste heat.
            They also have laundries and guest hot water which need a lot of heat.
            How stupid not to combine them?

            There is no financial incentive for the hoteliers

            If there exists a law that incentivizes small and medium business enterprises to re-use the excess heat they generate, I'm sure many would make the necessary changes

            Actually, a similar law has already been in effect for decades for heavy industries - factories that generate excess heat were encouraged to generate electricity and then sell the electricity to the power company

    • You'd think we'd have nanochannels by now.
    • by nbvb (32836)

      Does it come with Reference USB Keys?

      Sadly, I don't think anyone else reading slashdot was alive during the Micro-Channel era, let alone gets the joke ....

  • by mwfischer (1919758) on Monday June 18, 2012 @06:32PM (#40364477) Journal

    Hot and cold isn't that hard. Maybe I am missing a point somewhere.

    You take the heat energy biproduct from a processor and dump it somewhere else. In "normal" this case, an air conditioned room. Heat dissipated is being countered by air conditioning going 24/7. More energy.

    Instead of watercooling, which can refrigerate a fluid (more energy in put and unusable for anything else), this removes waste heat and reuses it elsewhere.

    This isn't going to work but... Instead of sitting in the tub and pissing in it continuously, your waste is being used somewhere else. (This is Germany after all)

    Has anyone ever tested if we actually need air conditioning for a server room? I mean transform one into a "wind tunnel" where the waste heat is either ejected outside or used internally? Instead of a giant cube... what about a rectangle?

    Will this lead to.. yo dawg I heard you like blades so we made your rack of blades into a blade?

    • by Anonymous Coward on Monday June 18, 2012 @06:39PM (#40364553)

      You ever try working in a server room when the a/c is broken? In southern California? In the middle of summer? It gets unpleasant. Quickly. And when it's 110 outside, you cannot simply pipe in outside air to cool the place. a/c is also for the server admins, not just the hardware. I'd venture to guess than the equipment will fail long after I have when the place gets a bit roasty.

      • by Bigby (659157) on Monday June 18, 2012 @06:57PM (#40364671)

        He didn't say "just get rid of AC". He was wondering if you designed the shape of the room where it has a constant draft. That way, the heat is cycled out of the building and cool air is pulled in from the other side. If you had a sever room that was 10 feet wide and 200 feet long, you could have one heck of a wind tunnel effect.

        • Re: (Score:3, Insightful)

          by pedrop357 (681672) *

          The hard part is where do you pull FROM. If it's like the poster said and 100+ outside, you're 'cooling' your servers with 100 degree air, and it's a convection oven for everyone inside that room.

          If you're pulling in the air from an air conditioned part of the building, you're just 'stealing' cold air from that part of the building and pulling it through your servers; you're also losing a lot of that cold air around the servers unless you isolate your hot/cold sides like some colo facilities do. Now we're

          • by ShanghaiBill (739463) on Monday June 18, 2012 @08:33PM (#40365351)

            The hard part is where do you pull FROM. If it's like the poster said and 100+ outside, you're 'cooling' your servers with 100 degree air,

            If the surface temp of your CPU is 160-200F, then cooling it with 100F air will work fine. You still have a delta-T of 60-100F. Computers do not need to be cooled with air that feels cool to a human. If the air temp is warm, it is usually much cheaper to increase the flow rate than to cool the incoming air.

            • by pedrop357 (681672) *

              Assuming all parts could handle the 100 degree air and the pressure that would be required. The room would be hell on earth with insane wind speeds, definitely a convection oven.

              • Assuming all parts could handle the 100 degree air and the pressure that would be required. The room would be hell on earth with insane wind speeds, definitely a convection oven.

                Do your maintenance at night, or early morning. In SoCal summer daytime temps are 100F+, but at night the temps are in the 70s.

          • by Anonymous Coward
            Most hardware can run just fine at 100+ degrees (I think most processors operate between 140F-180F). The problem is the equipment generates more than that. It's not just a matter of getting cold air (cold is relative) to the equipment as it is getting heat away from it. It starts with heat sinks which immediately helps dissipate the heat into the surrounding environment (ie the server room).

            Air is just a shovel for moving the heat away. Cooling the air just makes the shovel bigger so it can sit ther
        • I wonder how a chimney style might work. Pull 'cooler' air from the sides and eject hot air out the top.
          • by dbIII (701233)
            Perfect in cold places. Not so good for the rest of us, but there are interesting house cooling ideas like drawing the air in through long underground pipes so the incoming air is lower than ambient temperature.
            Where I am there is about month or two each year when the maximum outside temperature is lower than the 22C I have the server room set to. On the other hand, if it was a bigger room I could put an industrial sized solar airconditioning unit because there's a lot of sunshine here even in midwinter.
            • by cynyr (703126)

              I wonder if you can get enough "pressure" to pull the air though a direct evap cooler (DEC, Swamp cooler) before making it to the servers. There are many many places in this world were the humidity is low enough to make this work every day all day if you can accept 95-100F air into the servers, even when it is 120F outside.

              The bigger issue is not how to cool the air, but all the shit in the outside air. getting the air though filters and then the DEC with enough flow to make it work at 100F usually means a

        • by Artifakt (700173)

          What If you build a server room that's either centered around a vertical pipe, or shaped as your long and narrow design, but adding an uphill gradient? Maybe instead of a constant slope, you could start off flat at the intake end, and end up swooping upwards?
          On the other hand, how much of a problem is a constant stiff breeze in a server room? You could get a lot of cooling with a constant 35 Kph vertical wind, but can your techs work in one?

        • by Solandri (704621)
          The Romans used something like that [i4at.org] to cool their homes. The room itself doesn't need to be long and skinny (though it helps). All that really matters is that air enters at one end, and exits at the other. You run the incoming air through a long underground pipe which cools the air using the ground (which stays fairly cool in hot weather). Heated air at the other end naturally rises out a vent on the ceiling, drawing in cool air from the pipe. Basically it's like a passive geothermal heat exchanger.
        • Are you planning on pulling the air 10 feet across the room, or 200 feet along the length of it; 'cause if it's lengthwise, then heaven help your heatsinks at the air exit end of the room. 190+ feet of racks heating up the air. Nice wind tunnel.
      • by cnaumann (466328)

        You could still cool the servers with treated outside air (or 110deg water or whatever). You then insulate the server racks and cool the room with conventional AC. The trick then is that the AC only needs to handle the heat from the servers that leaks through the insulation rather than the power being dissipated by the servers.

    • by amorsen (7485)

      Search for "free cooling" and you will get the answers for your questions.

      • by afidel (530433)
        Yep, or "air side economizer". The biggest issue is you need to pick your site well as you need air that is neither too moist nor too hot for most of the year, and you really need to design it in as I've never seen a positive ROI on a retrofit. Intel ran some test unit for like 18 months to prove out that it could be done, but that's too short for most enterprises that run their gear for at least 48-60 months, though it could work for the web guys who tend to turn over their gear more quickly.
        • by cynyr (703126)

          Too moist? Servers will happily work with 95% humidity as long as it doesn't condense (which is easy as the server will heat the air and thus lower the RH), when it is cool out there is no reason to evaporate water to cool the air.

          (Yes I plot it on a psych chart if you would like).

          Lots of data centers are going to air side when the outside air will allow, and using other cooling methods when the air is too moist. This gives you more site flexability will still taking advantage of the "free cooling" with out

    • by TheLink (130905)

      Has anyone ever tested if we actually need air conditioning for a server room?

      You do need "air conditioning", since you do want to make sure the air is not too dirty or humid or dry or hot.
      But yes you can do without conventional data center air conditioning:
      http://www.facebook.com/note.php?note_id=10150148003778920&_fb_noscript=1 [facebook.com]
      http://www.datacenterknowledge.com/the-facebook-data-center-faq-newest-page/ [datacenterknowledge.com]

      They're also trying in a warmer more humid area:
      http://www.wired.com/wiredenterprise/2012/04/facebook-data-center-2/ [wired.com]
      Wonder how well that will work.

    • Yes, people have tested passive cooling ("wind tunnel") in server rooms. Integration density nowadays is way too high. In our (modest) server room CPUs and GPUs will quickly jump from die temperatures of 70C to 90C, which essentially means that the HW will be fried. :-/

      The problem with simply "dumping the waste heat" somewhere is that you need to find a place where to dump it. As the story indicated, this is not so much a problem in winter, but in summer, when no one wants your heat (no offices b/c sweat

    • by bryan1945 (301828)

      There is some company, whose name I of course forget at the moment, that is building a building aligned with the local persistent winds so that it does create a type of wind funnel affect. (That was the most awkward sentence I've ever constructed.) Somewhere in the northern midwest US, I think, maybe.

    • by plover (150551) *

      Yes, data center cooling has been studied extensively. Consider a typical data center with 5,000 square feet of server space, and wired for 145 watt-hours/sq ft. That center is drawing about 3/4 megawatts per hour. You don't just "guess" when you already know you have to get rid of 2.5 million BTUs of waste heat. You have to carefully plan the thermal flow through the data center to ensure the whole building doesn't simply cook itself in its own waste. That means using a combination of ambient air, wat

  • by Narnie (1349029)

    In related news, the UN Security Council has scheduled the start of WW3 to coincide with the 5 year anniversary of SuperMUC's online-date, pending the lack of adequate air-conditioning in the angry German scientists' offices.

    • by TWX (665546)
      Pretty much what I was thinking... When you get angry and design mechanical things, you start thinking of mean mechanical things.
  • by busyqth (2566075) on Monday June 18, 2012 @06:33PM (#40364493)
    Those scientists wouldn't be angry to have heated offices in the summer. Germany can be downright chilly in the summer. I remember some beautiful July days in Berlin with highs in the 50s.
    On top of that, heated offices will make the German scientists think they're in Mallorca or Costa del Sol and they'll be partying all day and night to the hot techno beats.
    • by steelfood (895457)

      If the cooling water comes out hot enough, it can also be used to heat the building's water supply. Or at least it can be used to help out, e.g. keeping the water in the pipes warm.

    • by gentryx (759438) *
      While that is basically true, I must still add that those offices are occupied by computer scientists. And those are not really known for partying hard, aren't they? And yes, I'm a CS PhD student, too. :-/
  • Seen on Slashdot (Score:5, Informative)

    by 1sockchuck (826398) on Monday June 18, 2012 @06:36PM (#40364521) Homepage
    Slashdot began tracking this one two years ago [slashdot.org].
    • "Tracking": is that our new euphemism for dupes? "Slashdot tracked eight stories this week." Sounds classy. I like it!
  • by pubwvj (1045960) on Monday June 18, 2012 @06:38PM (#40364541)

    "Rather than simply being dumped, SuperMUC's waste heat is designed to be converted into building heat during winter. Presumably it is mostly radiated away in summer"

    They might be storing the heat rather than dumping it in the summer.

    We are building a meat processing facility. Meat processing facilities use a lot of energy for heating water, cooling carcasses, freezing and general storage & air conditioning. To reduce our energy needs we're storing winter in thermal mass so that we can use it during the warm seasons. We're also using the 'waste heat' from our refrigeration compressors to heat water in addition to solar hot water and the backup of propane heating for the water. All of this will save us enormous amounts of money since we won't have to buy as much energy. Good for our carbon foot print and even better for our bottom line as more money will stay in our pockets rather than being dumped into the environment. IBM could do the same.

    See http://sugarmtnfarm.com/butchershop [sugarmtnfarm.com]

    • Talk about one-stop shopping, now I can get my high-energy particle simulations done AND get a nice ribeye roast without making separate stops...FINALLY!
      • by pubwvj (1045960)

        Shh... don't reveal our future technology projects! And rest assured, we've almost got the radiation levels of the ribeye down to acceptable levels post cooking in the collider. :)

  • by saibot834 (1061528) on Monday June 18, 2012 @06:58PM (#40364689) Homepage

    My university building is 80m from SuperMUC; there is a large campus at the site with several thousand students and employees. In winter it most definitely makes sense to use the heat from SuperMUC, as the average temperature is about 0 degrees Celsius. In summer it might be a bit more difficult to dissipate heat on hot days, though the average temperature is still only 19 decrees Celsius for July.

  • This is very simple water cooling. The principle is identical to what is found in high end overclocked systems.

    Your coolant only needs to be cooler than the core itself to remove heat. It's been known for a long time that dumping the heat of an overclocked system into a room through a water loop will heat said room.

    Even in the dead of winter when it's 0C outside, my *one* overclocked computer can keep my 300SQ ft room heated to above 70 degrees with no additional heat sources.

    News? I guess. Definitely a

    • by Anonymous Coward

      Non-overclocked air cooled systems also heat the room. This article is about heating rooms that the computer isn't in.

    • by Junta (36770)

      The very rough idea is obvious. Going through the specifics of instrumenting a facility, of determining what the acceptable temperature and flow rate are to keep cpu die temperature at an acceptable level (note, if your cpu is still kicking, that may not be enough, voltage leakage increases with temperature, meaning power draw goes up, and you are being inefficient by letting the die get *too* hot. Also, this is the fastest x86 based system in the world. In part because the cooling is adequate to let the

    • by bws111 (1216812)

      No, the coolant does not just need to be cooler than the core itself. It has to also be able to absorb heat faster than the core is generating it. This means that all the thermal resistances in the path must be accounted for. Usually this means that the coolant temperature must be MUCH lower than the core, which requires chilling the coolant. The difference with this system is that the incoming water temperature does not need to be particularly cool - it can be 45C/113F, which means it takes less cooli

  • BFD. I've been heating my house with computers for a couple of decades.
  • Respect the H2O (Score:4, Interesting)

    by MarcQuadra (129430) on Monday June 18, 2012 @07:03PM (#40364745)

    People need to understand and respect just how awesome water is as a coolant. The specific heat of the stuff (basically, how much heat you can 'sink' into a gram of it) and its benign, well-understood nature, and the fact that its density only changes a little bit between freezing and boiling points make it quite awesome.

    I live in a city with a river through it. I really don't know why they aren't doing cooling via air-to-water heat pumps. It's really absurd to blow fans all day when the river could carry away 100X the heat without too many ill effects.

    • by jgfenix (2584513)
      You would decrease the oxygen solubility in water and that would be bad for life in the river if you heated it too much.
    • I live in a city with a river through it. I really don't know why they aren't doing cooling via air-to-water heat pumps. It's really absurd to blow fans all day when the river could carry away 100X the heat without too many ill effects.

      Except for maybe killing most the marine life :P

      • by afidel (530433)
        What you can do is use the flow of the municipal water source as a heat dump. There are several places around the great lakes doing this since their supply water is in the 35-40F range year round and it's just going to heat up to ground temp on the way to the users anyways.
    • It's complicated (Score:3, Informative)

      by Anonymous Coward

      Water can be a b***h to use in a closed-loop cooling system. If it has any appreciable electrical conductivity, you get electrolytic corrosion of different metals in the cooling loop. If you use 18 megohm DI water, you get corrosion for other reasons. Depending on whether you have exposure to air (like in an evaporative cooling tower), you get bacterial and algal blooms, dirt, dust, pigeon poop - it's not as simple as "pump the water around in a circle and move the heat with it". Many closed loop water cool

    • by evilviper (135110)

      I live in a city with a river through it. I really don't know why they aren't doing cooling via air-to-water heat pumps. It's really absurd to blow fans all day when the river could carry away 100X the heat without too many ill effects.

      Closed-loop, ground-source heat-pumps are a bit more efficient than an open-loop water-source like you describe, and there's no concern about turning your cool and clear pristine rivers into a warm, stagnant swamp devoid of animal life.

  • by amorsen (7485) <benny+slashdot@amorsen.dk> on Monday June 18, 2012 @07:05PM (#40364775)

    It would be trivial to upgrade 45C water to e.g. 60C with a heat pump. This could be done with high efficiency, certainly COP > 3. Of course an office building might not need that much hot water in summer (maybe for showers for those who bike to work?), but other buildings nearby might. Or use it for district heating, if they have that in the area, but with existing systems that would probably require more like 70C.

    Or just get the CPU's running at 300C, of course. Then you could run a steam turbine on the coolant...

    • Yes this is interesting. Why cant we have components that stands for 110C then we could use the computers as water boilers and get steam out for heating or turbines.
  • Yawn... (Score:4, Informative)

    by Cyrano de Maniac (60961) on Monday June 18, 2012 @08:23PM (#40365289)

    Decades ago Cray heated their building in Mendota Heights Minnesota entirely using waste heat from the supercomputers. When they built their new campus a few miles away and sold the old building they had to go through some amount of trouble to retrofit it with heating from conventional fuels.

    What's old is new again.

The universe seems neither benign nor hostile, merely indifferent. -- Sagan

Working...