Slashdot Mirror


Keeping a Data Center Cool on the Cheap

jedimaud writes "You've heard of bubble wrap, and the boy in the bubble -- now, here's a datacenter in a bubble. I work for a government agency that, like most, is trying to cut back some costs, and one of those costs is a REAL datacenter. So, we decided to wrap the whole thing in plastic (including two 1.5 ton ACs). The room hovers about 83 degrees, however, the racks in the bubble (ok, more like a termite tent) stay about 10 degree cooler. Here's some pics to check it out."

319 comments

  1. Wow... by Computer+Mutt · · Score: 3, Funny

    Now that's cool.

    1. Re:Wow... by sgant · · Score: 2, Funny

      What's the last thing you hear right before a site is totally slashdotted?

      Here's some pics to check it out.

      --

      "Leo Fender was in a 'state of grace' when he designed the Stratocaster." -- Paul Reed Smith
    2. Re:Wow... by rootofevil · · Score: 2, Funny

      if servers get slashdotted, and theres nobody around to hear, do they really melt?

      --
      turn up the jukebox and tell me a lie
    3. Re:Wow... by zenbanana · · Score: 1

      Yeah, that's the pseudogeek equivalent of a guy in a jacked-up rusty pickup truck shouting "Hey, y'all, watch this!"

      --
      In theory, theory is better than practice, but in practice, it isn't.
  2. Hmmm by Captain+Splendid · · Score: 4, Funny

    Well, if your site is hosted by that Data Center, it just got a lot hotter in that bubble!

    --
    Linux, you magnificent bastard, I read the fucking manual!
    1. Re:Hmmm by nokilli · · Score: 2, Funny

      Timothy could post a dupe, and that would divide the heat by half, right?

    2. Re:Hmmm by aznrocket · · Score: 1

      For real ... it seems the /. effect has flooded the data center.

      The pictures link "contains no data"

      =/

  3. Wonder if... by ericdano · · Score: 2, Funny

    Wonder if the bubbles are working now that it's been SLASHDOTTED!!

    --
    It's either on the beat or off the beat, it's that easy.
    I moderate therefore I rule!
    --
    1. Re:Wonder if... by Stephen+Samuel · · Score: 1

      I'm guessing that the plastic is melting by now...

      --
      Free Software: Like love, it grows best when given away.
    2. Re:Wonder if... by schestowitz · · Score: 1

      Is it just my imagination, or is this warpping going to warp up the server? It's the blanket effect.

      If one of these self-temprature-controlling containers (like the astronauts use in space) gets used, does it really save energy?

      --
      My Linux - (L)ove (I)s (N)ever (U)tterly eXPensive
    3. Re:Wonder if... by shawb · · Score: 4, Insightful

      I think in this case the AC only has to work on pumping the excess heat from the servers out, and not cooling the whole room. Doesn't matter how warm the room gets outside of the bubbles.

      --
      I'll never make that mistake again, reading the experts' opinions. - Feynman
    4. Re:Wonder if... by JS_RIDDLER · · Score: 1
      --
      _JS
    5. Re:Wonder if... by Stephen+Samuel · · Score: 1
      I guess it depends on where you are, and what the weather's like at the moment. This time of year, it's likely as not to be at least resonably warm in most any US state (other than parts of Alaska). Come winter time, however, it could be advisable to start pumping in outside air and/or using the heat of the machines to warm the building.

      On the other hand, and not having seen the pictures, It's got me wondering if the data centre wasn't designed by someone a bit short on physics, who never realized that it's a bad idea to let the waste heat from your AC units stay in the space you're trying to cool.
      (( I know it's unlikely, but we've all read the Darwin Awards... ))

      --
      Free Software: Like love, it grows best when given away.
    6. Re:Wonder if... by Anonymous Coward · · Score: 0

      Captain, some neanderthal put a blanket on the server and it's all warped up. There'll be no warpping until further notice.

      -Scotty out

    7. Re:Wonder if... by Leroy_Brown242 · · Score: 1

      I did get a partial mirror up. I think I'm missing some pages though.

      http://forumpics.info/cooling/


    8. Re:Wonder if... by devilspgd · · Score: 2, Insightful

      Sure it matters -- The bubble still needs to insulate and the AC still needs to dissipate, both of which are greatly affected by the heat difference inside and outside the bubble.

      --
      Give a man a fish, he'll eat for a day, but teach a man to phish...
    9. Re:Wonder if... by Nutria · · Score: 1

      Come winter time, however, it could be advisable to start pumping in outside air and/or using the heat of the machines to warm the building.

      The dinosaur pen (which now also has loads of Unix, VMS & NT servers, since the m/f now only takes up 1/20th of the space it used to) at the company I work for is just north of NYC, and back in the days of huge disk farms of 3390s, they acutally did vent in winter air.

      --
      "I don't know, therefore Aliens" Wafflebox1
    10. Re:Wonder if... by ReallyNiceGuy · · Score: 1

      3 degress of separation from Vladimir Putin.

    11. Re:Wonder if... by Anonymous Coward · · Score: 0

      Right, so who needs insulation in their house? People must be stupid, their AC just has to pump the excess hea out, it doesn't matter how hot it is outdoors.

    12. Re:Wonder if... by shoemakc · · Score: 1

      Plus, occassionally you do need to ::::work::::: in a datacenter, and i can imagine myself being all that productive in a datacenter pushing 110 degrees.

      -Chris

      --
      --an unbreakable toy is useful for breaking other toys--
    13. Re:Wonder if... by pfafrich · · Score: 1
      I think in this case the AC only has to work on pumping the excess heat from the servers out, and not cooling the whole room.

      Finally a use for all the AC's who post to slashdot!

      --
      There are four sorts of people in the world: fools, lunatics, idiots and morons. - Umberto Eco, Foucaut's pendulum.
    14. Re:Wonder if... by Anonymous Coward · · Score: 0

      As a regular slashdot reader, I'm getting increadibly pissed off with these kind of posts. This site is supposed to be a news site, not some kind of sick entertainment to see how quickly the community can bring down a server.

      Comments like "Wonder if the bubbles are working now that it's been SLASHDOTTED!!", give this community a bad rep, and should be modded down or damn well removed.

      And people wonder why slashdotters are arrogant, anti-M$-script-kiddies who get off on bringing down a site.

      Get a life.

    15. Re:Wonder if... by BK425 · · Score: 1

      More importantly, in terms of AC, "only has to work on pumping the excess heat from the servers out, and not cooling the whole room" doesn't make a whole lot of sense. The size of the space isn't as important as the heat load. Put another way, that empty space isn't making any heat, it doesn't really present any work for the AC to do.

    16. Re:Wonder if... by devilspgd · · Score: 1

      Probably, although if you have heat being pumped into the server room (from a furnace), it might be worth the effort to not spend too much time cooling the hot air being pumped into the room.

      (Again, I'm thinking extremely low budget, where blocking the heat isn't feasible)

      --
      Give a man a fish, he'll eat for a day, but teach a man to phish...
    17. Re:Wonder if... by BK425 · · Score: 2, Interesting

      Sorry, I don't understand. Are you saying there's a server room with a (non pentium) heater going into it? I've never seen anything like that. I would imagine that if it's not on a seperate zone (ie it's own thermostat) you could just block the airflow from the furnace to that room and have an HVAC guy rebalance the airflow to other vents in the zone.
      But, I've never seen a room intended for servers, that had an actual furnace attached to it.

      Other posters on this thread have suggested that having datacenters in Siberia or Alaska might be a good idea. This is not new. Alaska advertises several secure datacenters that were started with public bonds. Uninteruptible power takes on a whole new meaning when you're within pipeline distance of a natural gas field and wire distance of a coal plant. And at least for the one that I read about the entire facility was normally heated from the heat stripped off the server spaces by airconditioning (heat pumps). Connecting a heater to a server room would be a bad idea imo.

    18. Re:Wonder if... by devilspgd · · Score: 1

      I've seen older buildings with a server room that happens to not have it's own thermostat.

      With forced air you can usually block it. With radiator-style heating, if the water needs to pass through the server room to an adjacent office, you can't block or disable the heating in any way.

      --
      Give a man a fish, he'll eat for a day, but teach a man to phish...
    19. Re:Wonder if... by BK425 · · Score: 1

      I imagine that -might- happen, but this is starting to seem esoteric to the point of... unlikeliness. Fwiw I work for a 20 person company in Seattles Pioneer sqaure district in a brick building on the historic roster... our radiators were removed on the verge of failing I think, probably 10 years ago.

      At home I have hot water baseboards that I routinely block with styrofoam, with steam radiators maybe you couldn't do that. At some point though, all of this neat "what could we do if..." and "lets find the cheapest solution for..." comes down to: Will you burn up 20k or more worth of servers to avoid calling someone with expertise in heating systems?

      I'm not -really- that interested in HVAC. The Mech E's who are, have experience and skills that are worth the money that businesses pay for them. I originally posted because I thought the relation to computing was interesting, but... if you -really- put a server room in a building with valveless steam radiators... I don't think your problem is amenable to "hacking" on the cheap. Or in the long run even related much to your heating system. Certainly plastic painters sheets aren't going to solve it ; )

      That's really it for me, I promise to post no more forever on slashdot about HVAC engineering. (And tomorrow I scout out how to post privately to one person.) BK425

  4. Warning: Suffocation risk by dangitman · · Score: 5, Funny

    Do not place datacenter over head. Keep out of reach of children.

    --
    ... and then they built the supercollider.
    1. Re:Warning: Suffocation risk by Mister+Impressive · · Score: 5, Funny

      Do not eat datacener.

      --
      Let the commencement BEGINULATE!
    2. Re:Warning: Suffocation risk by YouCanCallMeAl · · Score: 1

      Do not taunt datacenter...or is that Happy fun ball?

    3. Re:Warning: Suffocation risk by modecx · · Score: 1

      No, I'm pretty sure that also applies to datacanter, especially if it belongs to the government, and even more especially if it belongs to the NSA!

      --
      Constitutional rights may be respected, repealed, or modified; but they must never be ignored.
    4. Re:Warning: Suffocation risk by Inominate · · Score: 1

      Do not taunt datacenter.

    5. Re:Warning: Suffocation risk by Knara · · Score: 4, Funny

      Do not taunt the happy datacenter

    6. Re:Warning: Suffocation risk by HG+Slashdot · · Score: 1

      Do not slashdot a datacenter.

      --
      j0b.org - A famous domain name for sale
    7. Re:Warning: Suffocation risk by Anopheles · · Score: 1

      If Datacenter begins to smoke, get away immediately. Seek shelter and cover head.

    8. Re:Warning: Suffocation risk by NothingCleverToSay · · Score: 1

      Happy datacenter may stick to some types of skin.

  5. Needs more duct tape... by tinrobot · · Score: 3, Funny

    Because... stuff always needs more duct tape.

    1. Re:Needs more duct tape... by Seumas · · Score: 1

      Because... 4mm thick sheets of plastic insulate really well.

    2. Re:Needs more duct tape... by Leroy_Brown242 · · Score: 4, Funny

      and more cowbell.

    3. Re:Needs more duct tape... by Anonymous Coward · · Score: 0

      duct tape + 2( 1.5 tonne AC's ) = not a pretty site. Think se7en dinner party.

    4. Re:Needs more duct tape... by theTerribleRobbo · · Score: 1
    5. Re:Needs more duct tape... by HABITcky · · Score: 1

      And more jiggawatts.

  6. Note from the metrology freak by Guppy06 · · Score: 5, Informative

    "(including two 1.5 ton ACs)"

    "Ton" here refers to a "ton of cooling," a measure of power. It was originally intended to mean "the power required to melt one ton of ice in 24 hours." Since that varies based on a bunch of conditions, it was pegged at 12,000 Btu/h.

    When they changed the definition of "calorie" to mean 4.1868 J, converting Celsius to Fahrenheit and grams to pounds gives us a conversion factor of 1 ton of cooling being exactly 3 516.852 842 066 7 W.

    In other words, each a/c unit is about 5.25 kW of cooling each, or 10.5 kW total.

    Oh, and 83 degrees Fahrenheit is about 301 kelvin and a ten-degree Fahrenheit difference is a difference of 6 kelvin.

    (According to my old HVAC prof, there's been little to no progress in "metricizing" the industry in the US. Having used both systems in his course, I'd say I prefer US units, if only because the unit descriptions on things like insulative properties make more sense when the units for thickness and area don't naturally cancel each other out.)

    (And it could be worse. Most home a/c units are labelled on the box as putting out x number of Btu, suggesting they're disposable.)

    (Well, they probably are...)

    1. Re:Note from the metrology freak by Solder+Fumes · · Score: 5, Funny

      That was entirely tangential and ultimately useless information that adds no value to this discussion. ...which means that it was perfect for Slashdot! Keep up the good work!

    2. Re:Note from the metrology freak by Guppy06 · · Score: 1

      Meh. There's a reason why I turned off the bonuses, and I decided to respond to the usual "US units are t3h suxx0rs! /. is too US-centric!" bitchings before they started. :)

    3. Re:Note from the metrology freak by Anonymous Coward · · Score: 0, Flamebait

      Somebody learned about 'tons' in class this week. I bet the teacher loves his special student.

    4. Re:Note from the metrology freak by The_Wilschon · · Score: 1

      erm... hate to burst your bubble, but last time I checked, a nine degree Fahrenheit difference was a difference of five degrees Celsius, or five Kelvins. If the ratio were 10:6, then, assuming that 273 K = 32 degF, then 373 K would be 198 degF. Which is not correct. (The temperatures chosen were freezing and boiling of water at 1 atm pressure, in case anyone didn't know. The correct Fahrenheit boiling point of water (at 1 atm) is 212 degF.)

      --
      SIGSEGV caught, terminating

      wait... not that kind of sig.
    5. Re:Note from the metrology freak by Anonymous Coward · · Score: 5, Funny

      (including two 1.5 ton ACs)

      Hey, I may be a fat geek, but I'm not THAT fat!

    6. Re:Note from the metrology freak by Guppy06 · · Score: 4, Informative
      " but last time I checked, a nine degree Fahrenheit difference was a difference of five degrees Celsius,"

      ... which means 10 is 5 5/9, rounding up to 6. Guess I didn't use the word "about" enough times for you.

      But hey, if you really wanna get anal, do it right.

      "assuming that 273 K = 32 degF,"

      32 degrees Fahrenheit is 273.15 kelvin.

      "The temperatures chosen were freezing and boiling of water at 1 atm pressure, in case anyone didn't know."

      At 101 325 Pa, water boils at 373.124 kelvin. Celsius /= centigrade. The freezing point is also flakey, but not within three decimal places at least. This is why all modern temperature scales are pegged to the triple point (273.16 K)instead of trying to correlate two data points that move around with respect to each other.

      "The correct Fahrenheit boiling point of water (at 1 atm) is 212 degF.)"

      Try 211.953 degrees Fahrenheit.

      Oh, and the first letter of SI units (like kelvin) is never capitalized.

    7. Re:Note from the metrology freak by l3v1 · · Score: 1

      converting Celsius to Fahrenheit and grams to pounds gives us

      Eek, I wouldn't do that, ever.

      I'd say I prefer US units

      Of course you prefer. Me, I would fire every engineer who would not use the metric system for calculations. If it's good or not that can be [was,is] a subject of debate, but thing is, it works, and most of the world knows it. IMO the best way is to use metric units and convert to others if and only if necessary. Very many troubles can be escaped this way.

      --
      I am putting myself to the fullest possible use, which is all I can think that any conscious entity can ever hope to do.
    8. Re:Note from the metrology freak by BlackMesaLabs · · Score: 0

      ...Maybe he meant the weight of the AC unit like everyone else thought? :P

    9. Re:Note from the metrology freak by Anonymous Coward · · Score: 1, Funny

      Go stand in the corner and think about what you just said.

    10. Re:Note from the metrology freak by William+Robinson · · Score: 3, Funny

      He must be a consultant!!:)

    11. Re:Note from the metrology freak by Anonymous Coward · · Score: 0

      > or 10.5 kW total.

      Good lord... that's hardly anything. That rating must assume (all jokes aside) 100% efficency. Japanese aircons barely able to keep 10 tatami mats worth of space cool (and remember, we're talking small human beings as the primary heat source) are rated at 2.8kW.

      Either that... or his "datacenter" fits in a largish walk-in closet.

    12. Re:Note from the metrology freak by Anonymous Coward · · Score: 0

      Obviously not the smartest Skittle in the bag, are ya?

    13. Re:Note from the metrology freak by Sinner · · Score: 1
      ... a conversion factor of 1 ton of cooling being exactly 3 516.852 842 066 7 W.
      I'm sorry, this is Slashdot. Could you give me that in standard "Libraries of Congress" units?
      --
      fish and pipes
    14. Re:Note from the metrology freak by Guppy06 · · Score: 2, Insightful

      "Eek, I wouldn't do that, ever."

      Then you're limiting you're usefulness in the workplace. Even if everything in the US (and exported to the US) was manufactured to SI specifications tomorrow, there's still older equipment to maintain. And if you refuse to figure out what size metric socket is needed to turn a 1/2-inch bolt because it's against your religion, that's one less application for HR to consider.

      "Me, I would fire every engineer who would not use the metric system for calculations."

      Straw man. I said in the sentence immediately before your selective quote that I used both in that particular course, just like just about every student in every engineering course in every US school.

      "but thing is, it works,"

      So do US units, to exactly the same degree. US units have been defined in terms of SI units since the Nineteenth Century.

      The difference is in application. SI units tend to be tied to certain hidebound rules of use and stigmas that aren't attached to US units.

      Power is always measured in watts, whether we're talking about cooling capacity or electricity. So does this mean if I plug a 5.25 kW air-conditioner into the wall that it will pull 5.25 kJ of heat out of the air per second, or that it will consume 5.25 kJ of electricity per second, or (even worse) both? In the US, you're allowed to use tons for cooling capacity even while still using watts for electrical power.

      SI requires you to measure energy transfer per second, even if you're more interested in hours or days (which you are in HVAC). SI requires you to always use meters instead of centimeters in unit definitions (and cancel when able), leaving you with ambiguous units like "watt per meter-kelvin" instead of "btu-inch per hour-square foot-degree Rankine." When you're dealing with wall areas and insulation thicknesses and analyzing the heating and cooling requirements of a room over the course of hours, it's actually less work to use US units than SI units because of all the rules BIPM churns out on the proper use of SI.

      "Very many troubles can be escaped this way."

      Others are introduced by the insistence on one set of units over another. Was the Mars Orbiter lost because the manufacturer didn't use SI, or was it lost because NASA insisted on SI when every other customer in the aerospace industry insists on US units? The civillian world does things like measure airplane altitudes in units of 30.48 cm and ships cargo in containers 12.192 m in length because demanding that everything be done in SI would cause "very many troubles."

      SI vs US is ultimately arbitrary. Both have advantages and disadvantages. But demanding that everybody in the world use SI in all instances involves as much hubris as demanding everybody in the world speak English in all instances.

    15. Re:Note from the metrology freak by SavvyPlayer · · Score: 1

      Actually it was the first relevant comment posted, putting into perspective the efficiency and cost savings of this solution. Few datacenter managers and even fewer VPs have the HVAC training needed to appreciate "1.5 tons of cooling".

    16. Re:Note from the metrology freak by mjpaci · · Score: 1

      Actually, thank you. I was just having a conversation with an HVAC guy last weekend and he didn't exactly know what a ton of cooling really was. He estimated it at 12,000 BTU but didn't know what it referred to.

      --mike

    17. Re:Note from the metrology freak by bbtom · · Score: 3, Funny

      Too knowledgable for consulting. Probably an academic.

      --
      catch (HumourFailureException e) { e.user.send("You, sir, are a humourless idiot."); }
    18. Re:Note from the metrology freak by Anonymous Coward · · Score: 0

      *snap* *snap*
      You go girl!
      *snap* *snap*

    19. Re:Note from the metrology freak by Anonymous Coward · · Score: 0

      Now..... is that a metric or an imperial ton of ice?

    20. Re:Note from the metrology freak by Anonymous Coward · · Score: 0

      Nope, if he was an consultant he would be paid for this one, and/or he would recommend some expensive hardware/software to solve it.

    21. Re:Note from the metrology freak by rcw-work · · Score: 1
      That was entirely tangential and ultimately useless information that adds no value to this discussion. ...which means that it was perfect for Slashdot! Keep up the good work!

      Heh.

      Actually, converting from tons or BTU/h to kilowatts is very useful, because that's the native measurement of your heat load. The amount of heat you need to remove from a room is equal to the amount of heat you add, which is equal to the amount of electricity your equipment uses. If the ammeter readings on the circuits going to the room total up to, say, 50 amps, and that power is at 120 volts, and you'll probably have the room stuffed with double the equipment in a few years, then you need to plan on removing 12kW. However, that's greek to HVAC people, so tell them you need a 4 ton unit.

      (Immediately remove any HVAC contractor or consultant from the building that attempts to measure the area of the floor to determine equipment cooling needs. The room doesn't get hot on its own, the equipment does.)

    22. Re:Note from the metrology freak by Anonymous Coward · · Score: 0

      At 101 325 Pa, water boils at 373.124 kelvin

      More Kelvin trivia - you're supposed to use the plural form if you write the word:

      373.124 K or 373.124 kelvins.

    23. Re:Note from the metrology freak by marcosdumay · · Score: 1

      "I'd say I prefer US units, if only because the unit descriptions on things like insulative properties make more sense when the units for thickness and area don't naturally cancel each other out."

      That is because you learned that way. Anyway, the units don't cancel each other just because you can mix several different units that measure the same thing on the US units, try using just feet and square feet and you'll see them cancel (you can also replace ft/in by 12, yd/ft by 3...).

    24. Re:Note from the metrology freak by flappinbooger · · Score: 1

      yeah, but the room will also get some heat added roughly proportional to it's floor area if it is exposed to unconditioned space via the ceiling or floor or walls. Also, people give off heat and moisture too.

      It's tough to design HVAC adequate for data centers not only because you don't know future equipment needs, but it's tough to nail down PRESENT equipment needs during the design phase.

      (HVAC consultant. Sorry.)

      --
      Flappinbooger isn't my real name
    25. Re:Note from the metrology freak by vbrtrmn · · Score: 1

      Heh, US units are Metric, according to the Metric Conversion Act of 1975.

      --
      it's a sig, wtf?
    26. Re:Note from the metrology freak by dillon_rinker · · Score: 1

      HE is my hero...because he IS the Anti-Pedant!

    27. Re:Note from the metrology freak by Guppy06 · · Score: 1

      Neither. The US uses the "short" ton of 2000 lb. Everybody (US or otherwise) tends to write 1000 kg as "tonne" and the people who might have used the imperial ("long") ton of 2240 lb are no longer allowed to.

    28. Re:Note from the metrology freak by Guppy06 · · Score: 1

      "(you can also replace ft/in by 12, yd/ft by 3...)

      Yes, you can. The difference is that, in SI, you must. The SI standard for length is the meter and only the meter, and units that want to stay within the bounds of SI have to be based soley on the meter (and not the centimeter, lest you start using CGS units, which are just as frowned upon as feet and pounds).

      In US units, there's no rule saying that you can't use inches of wall thickness and square feet of wall area in the same unit name, avoiding the cancelling out of like units (unless you really want to invoke the conversion factor). There is no One True Unit of Length that must be preferred in all instances.

      Similarly, in energy transfer, in the US you can talk about Btu/s, Btu/h, ft-lb/s or ft-lb/h, using whichever measure of energy is more convenient to your purposes (thermal or mechanical energy) and measured over whichever time interval you find most useful. In SI, there is only W. All energy must be J (kilowatt-hours are not allowed by SI) and all time intervals must be s (other time units are allowed on their own, but not in unit definitions).

    29. Re:Note from the metrology freak by Anonymous Coward · · Score: 0

      Litre is capitalized.

      - a Canadian

  7. Real Data Center by aaarrrgggh · · Score: 4, Informative

    First off, a "real" data center needs a little more than 1.5 Tons of cooling; that will barely cool a single rack in a real data center.

    Insulation is always a nice idea, be the fact of the matter is that to reject the heat from the space you need to provide a means of heat trasnfer. Generally, that requires a temperature differential between your heat source and the outside. If it is cooler outside than in the space, not much is required. If it is warmer, you will have to take advantage of thermodynamics and use a compressorized cycle. This can be more or less efficient, depending on the difference in inside and outside temperatures.

    (A typical data center operates with a 95-110F outside design temperature, and attempts to deliver 48F chilled water to the CRAC units (Computer Room Air Conditioning). This forces about 50% of the energy consumed by the computers to be used (again) to cool the equipment.)

    Call an engineer when everything melts down...

    1. Re:Real Data Center by Craigj0 · · Score: 3, Interesting

      First off, a "real" data center needs a little more than 1.5 Tons of cooling
      I'm pretty sure 3 (2x1.5) is considered a little more than 1.5 but perhaps I'm not using "real" math.

    2. Re:Real Data Center by cyberfunk2 · · Score: 1

      Do you mean that the energy is reused or 50% additional ?

      He he... CRAC units.

    3. Re:Real Data Center by SuperQ · · Score: 1

      Yea.. two 1.5 ton units is nothing.. I have a 2.5 ton unit on my house, and it's more than enough to keep out the hot and humid we've had for the last week or so.

      the DC where I keep my colo has about 25 tons of cooling, to keep up with about 40kW of systems.

      (it's always good to have spare cooling capacity, especialy because you want to give the compressors a break so they arn't running 24/7)

    4. Re:Real Data Center by berndtj · · Score: 1

      Well, I don't think that this discussion is really about real data centers, but more for small businesses that have a few file servers or a single rack. Ideally there would be a small room just dedicated to the servers. That way the bubble would consist of insulated walls. The truth is though that air makes an excellent insulator, so just having a plastic barrior between the inside and outside of the server really isn't too bad. The biggest issue I think would be circulation. If anyone has spent some time in a data center, you will notice hot spots where the cool air doesn't reach. Also, generally the smaller the ac unit, the less efficient, so throwing both racks in a small room with one 2.5/3 ton mr. slim ac unit, would probably be a better solution. Lastly, I like duct tape engineering as much or more than most, but that looks ghetto!

    5. Re:Real Data Center by Anonymous Coward · · Score: 0

      And given than 1.5 tons is about 4.5 kW, or 37.5 amps at 120 volts, I wonder what sort of systems are in the racks at these "real" data centers. At my imaginary data center, the only time I'll see 4.5 kW per rack is if the rack is fully stuffed -- a rare occurance.

    6. Re:Real Data Center by aaarrrgggh · · Score: 1

      General practice is to provide redundant cooling, so I only counted a single unit. Them things need to get shut down for maintenance!

    7. Re:Real Data Center by Grendelll · · Score: 1

      I'd think that all of that cold water in my CRAC would speed me up significantly.

  8. Damn... by eldawg · · Score: 2, Funny

    I doubt anyone had a chance to cache these pics.

    1. Re:Damn... by Adrilla · · Score: 3, Informative
      --

      "Plans are for fools! Oglethorpe, the plutonian (Aqua Teen Hunger Force)
    2. Re:Damn... by dustinbarbour · · Score: 1

      Ya know.. I'm appreciative of MIrrordot for the service they provide, but sometimes it sucks. For instance, linking to a photo gallery without mirroring the fullsize images. :/ A little annoying, but hey.. thanks for what you do anyway!

  9. Tax money... by loony · · Score: 2, Funny

    At least now I know where my tax money went! Termite tents! That's at least more believable than that $15,000 toilet seat and $20,000 hammer :-)

    Peter.

    1. Re:Tax money... by fyndor · · Score: 1

      Want to know the reason why? Its because the govt requires an obsurd amount of testing for everything so crap like a toilet seat end of cost unreasonable amounts of money.

    2. Re:Tax money... by Anonymous Coward · · Score: 0

      No, not at all actually. Its because they require EVERYTHING to be listed with price. ( or however the hell you want to say it ) And most contractors for the gov have lists soooo damn long its mostly just glanced over. So contractors bump up prices on little things when they can, thus making money.

      1) Become Gov Contractor
      2) Charge $2 per screw and $3 per nail, etc, etc

      3) ......

      4) PROFIT LIKE NO ONE HAS BEFORE!!

      was obligitory.. common.

    3. Re:Tax money... by Anonymous Coward · · Score: 0

      I'd always assumed it was just a way of hiding non-public (ie black) defence appropriations.

    4. Re:Tax money... by Nutria · · Score: 1

      Want to know the reason why? Its because the govt requires an obsurd amount of testing for everything so crap like a toilet seat end of cost unreasonable amounts of money.

      When I worked for a defence contractor, contract prices correlated linearly with the thickness of the RFP.

      IOW, if civil service and/or Pentagon yahoos took the trouble to write up a 5 page spec defining exactly what kind of "hammer" they wanted, they got charged $200 for that claw hammer, since it took us so much time to verify that the $5 Stanley hammer actually met the 5 pages worth of specifications.

      --
      "I don't know, therefore Aliens" Wafflebox1
    5. Re:Tax money... by Dogtanian · · Score: 1

      IOW, if civil service and/or Pentagon yahoos took the trouble to write up a 5 page spec defining exactly what kind of "hammer" they wanted, they got charged $200 for that claw hammer, since it took us so much time to verify that the $5 Stanley hammer actually met the 5 pages worth of specifications.

      That as may be, isn't it also a convenient tactic for the US government to divert money and resources into Top Secret hush-hush projects that they really *don't* want people finding out about?

      --
      "Slashdot - News and Chat Sites Deviant". (Click "homepage" link above for details).
    6. Re:Tax money... by nolife · · Score: 1

      Sometimes the testing is worth it because there are MANY government suppliers that would supply absolute junk if they could. I worked on this in a much smaller scale with computer parts in the military. There was a local vendor that would always win every bid for parts we had for basic computer parts like hard drives, zip drives, mother boards etc... In the two years we dealt with this vendor, I don't think we ever received the correct parts or all working parts. The guy knew how the system worked and knew how to manipulate it. We would order 10 zip disks, we would receive 5, we got MBs with dead serial ports, mice with the serial cable wired incorrectly, dead floppy and HDs. It was comical. The amount of paper work overhead and the constant turnover of people we had made his junk supply system work.

      --
      Bad boys rape our young girls but Violet gives willingly.
    7. Re:Tax money... by Anonymous Coward · · Score: 0

      Hey! You saw Independence Day too?

      At least Judd Hirsch is keeping the military-industrial complex honest!

    8. Re:Tax money... by Thundersnatch · · Score: 1

      No, the real reason is that most of the $15,000 toilet seat money actually went into the "black projects" budget. That's how they built things like the F-117A and the B-2 stealh planes in secret, and keep them secret for so long. Even the DoD isn't stilly enough to have a budget line item reading:

      January 1987...$2,300,000,000...secret plane factory in Nevada
  10. Really a good idea? by novalogic · · Score: 3, Insightful

    If that A/C unit freezes up/dies/etc, getting wraped in that bubble will cause those machines to overheat rather quickly...

    Might be a good idea to hookup a tempature controlled moter to pull the plastic down if the A/C dies, if you know how, and have a good junk pile, you can do that cheap enough....

    --
    --
    1. Re:Really a good idea? by Anonymous Coward · · Score: 0

      hmmm, that is a really good point. Perhaps they could keep positive pressure in the bubble when the ac and the fans are running. Then have lightly springged flaps that would open when the pressure inside the bubble dropped. that would at least help to let the hot air out in an emergency.

    2. Re:Really a good idea? by HermanAB · · Score: 3, Funny

      Actually the takedown of the tents is automatic: When the computers overheat, the plastic will slump and drop onto the floor...

      --
      Oh well, what the hell...
  11. Build the datacenter in alaska by team99parody · · Score: 3, Interesting

    Clean dry cold air. Self sufficient oil. Great physical security

    1. Re:Build the datacenter in alaska by Solder+Fumes · · Score: 4, Insightful

      Heck, why not? Run some massive fiber along the big pipeline. Only disadvantage is that you're out on the edge of the Internet, you're not safely embedded in the middle with nice redundant capabilities.

    2. Re:Build the datacenter in alaska by Leroy_Brown242 · · Score: 1

      I don't know about now, but Alaska didn't used to have a very reliable power grid.

      Any benefits might be offset by having to have huge UPS and generator systems.

    3. Re:Build the datacenter in alaska by Anonymous Coward · · Score: 0

      Any datacenter that is mission critical has UPS and generator systems.

    4. Re:Build the datacenter in alaska by Leroy_Brown242 · · Score: 1

      Yes but in a place where they could go down for much longer, they would need much more.

      Making thigns expensive quickly.

    5. Re:Build the datacenter in alaska by amliebsch · · Score: 1

      And energy efficient, because the datacenter can heat your human housing and work quarters. Maybe you made this comment as a joke, but...it seems like it would be an ideal location for massive computing operations.

      --
      If you don't know where you are going, you will wind up somewhere else.
    6. Re:Build the datacenter in alaska by Solder+Fumes · · Score: 1

      The ideal location for massive computing operations would be in space. Not the ideal location for gaming servers, but a good place for supercomputers. Build a Beowulf cluster of satellites, send up more computer satellites to upgrade the cluster.

    7. Re:Build the datacenter in alaska by whmac33 · · Score: 1

      Not a bigger generator, just a bigger diesel tank. :)

    8. Re:Build the datacenter in alaska by Anonymous Coward · · Score: 3, Informative

      Excuse me, do you run a datacenter for a living?

      When you are running a mission critical datacenter (this is usually all of them, if we are talking about real datacenters here, not the closet of your 5000 square foot PC repair shop where you keep your web server), you have supplies to get you through any length of power outages. This usually includes supplies of diesel fuel and a contract with a fuel supplier to bring more diesel on site for the duration of the outage.

      Do you have any clue how major datacenters in metropolitian USA are run? You don't depend on the reliability of the grid to determine how to budget power systems for your datacenter.

      -AC with a clue

    9. Re:Build the datacenter in alaska by TheFlyingGoat · · Score: 2, Interesting

      Build it in a basement with a good vapor barrier. To make it more efficient to cool, use a cooling system that makes use of a ground loop of coolant. They're extremely efficient and not too expensive.

      Without the a/c on, my basement hovers around 68 degrees throughout the year. It's much easier to control the temperature than elsewhere in the house. With a way to actively dissipate heat (the ground loop), it'd be the ideal datacenter location without having to man a post in Alaska. :)

      --
      You have enemies? Good. That means you've stood up for something, sometime in your life. --Winston Churchill
    10. Re:Build the datacenter in alaska by Leroy_Brown242 · · Score: 1

      okay, AC. But putting a power reliant, mission critical, data center on an unreliable power grid, is stupid.

      So what if you can keep it on by feeding the generators more fuel. It's pretty cost inefective to rely on that very often.

      Especially when you easily put it on a more reliable network, and not spend money on deisel.

    11. Re:Build the datacenter in alaska by HyperChicken · · Score: 1

      Great physical security

      Security through obscurity isn't security at all.

      --
      Free of Flash! Free of Flash!
    12. Re:Build the datacenter in alaska by kfg · · Score: 2, Funny

      "Hey, how come we're down?"

      "Don't know. Must be a caribou in the works."

      (And when they track it down I want to see them tape that to an index card)

      KFG

    13. Re:Build the datacenter in alaska by Anonymous Coward · · Score: 0

      Don't forget those ICBM defense missiles..

    14. Re:Build the datacenter in alaska by amliebsch · · Score: 1

      What are you talking about?!? How would space be a good location for computers at all?

      --
      If you don't know where you are going, you will wind up somewhere else.
    15. Re:Build the datacenter in alaska by Anonymous Coward · · Score: 0

      Space, ideal? Not really. How is heat removed? Three major ways: radiation (not nuclear radiation, you nong), convection, and conduction.

      Convection relies upon the presence of air, so that's out. Conduction relies upon the presence of solid matter. Again, that's out. That leaves radiation.

      You're going to need some pretty serious surface area to keep the equipment cool enough for your needs. Then there's preventing that area from absorbing the heat radiation from the sun ...

    16. Re:Build the datacenter in alaska by Anonymous Coward · · Score: 0

      Actually, it could be potentially much more $$$ efficient to run your entire opreation off of generators. Bought in bulk--without the taxes you pay for road use, a couple tanker trailers would last a decent sized place quite a while, and might have the same net-effect of the grid, except being more efficient energy wise, pretty close cost wise, once you factor in redundancy, maintainacne, etc.

      What one needs is a couple locomotive engines tied to some heavy-ass 3 phase generators, located in a sound dampening box.

    17. Re:Build the datacenter in alaska by Leroy_Brown242 · · Score: 1

      That sure does sound ecologicaly friendly too! :)

    18. Re:Build the datacenter in alaska by atarione · · Score: 4, Funny

      the real problem with a data center in alaska is that the fucking polar bears will keep drinking all the cherry coke.

      --
      actually I am happy to see you, however that is in fact a banana in my pocket.
    19. Re:Build the datacenter in alaska by lachlan76 · · Score: 1

      No it wouldn't...radiation, as well as the fact that there is nowhere to dissipate the heat into would be a big obstacle.

    20. Re:Build the datacenter in alaska by barry99705 · · Score: 1

      Naa, the power is much better now.

    21. Re:Build the datacenter in alaska by mc6809e · · Score: 1

      The ideal location for massive computing operations would be in space. Not the ideal location for gaming servers, but a good place for supercomputers. Build a Beowulf cluster of satellites, send up more computer satellites to upgrade the cluster.

      Bad idea since the only way anything loses heat in space is through radiation.

      Just think of hot soup in a thermos. It takes a LONG time to cool down. That's cooling through radiation.

      Air is nice to have around here on Earth since it can carry the heat away must faster than it can be radiated away.

    22. Re:Build the datacenter in alaska by devilspgd · · Score: 1

      Actually it's not bad at all in terms of cost -- Slightly more then when you're running on the grid, but as long as you have the UPS capability to handle the switchover gracefully, it's not a big deal.

      Assuming you're buying in bulk (Meaning you need large storage capacity too) you can get the price to within a few percentage of buying electricity. The additional cost of running on generator power when needed will be offset by the fact that you don't need cooling. When the grid is up and running, you'll be pocketing that money.

      If nothing else you could probably heat the building housing your staff, maybe run a small call center out of the same building as the datacenter.

      --
      Give a man a fish, he'll eat for a day, but teach a man to phish...
    23. Re:Build the datacenter in alaska by Anonymous Coward · · Score: 0

      High energy protons spilling through the earth's magnetic weak spots causing massive spikes in any significant electrical circuits....

      Yeah... that's where I want to build *my* datacenter.

      And you do realize that it's only going to get worse, right?

    24. Re:Build the datacenter in alaska by kingofalaska · · Score: 2, Informative
      That depends on where in Alaska.

      Right now, in south central Alaska, it is 50 degrees outside. It was in the 80's earlier. I have worked on projects all over Alaska, and it gets hot and humid in the Brooks Range and Fairbanks areas, then drops to -40 for weeks at a time in the winter. Other areas are milder, but humid. I read that Anchorage is one of the sweatiest areas of the U.S., and I'm glad I don't live there. But it is Bikini season here, a ways north.

      As for the 'self-sufficient oil', most of our oil is sent to the small states, and we supply 25% of the U.S. oil consumption. OTOH, we have vast supplies of untapped gas, coal, and radioactive materials.

      And 'physical security'? Better look at a map. Alaska is not a small island off the coast of California. It's the air crossroads of the world, and within spitting distance of Siberia. It is the only part of the U.S. to be occupied by enemy forces during WWII. Of course, everyones grandma has a gun, and knows how to use it. And we have the greatest concentration of veterans of all the states. I pity the fool who attacks Alaska.

      KoA

      Navy to Test Shape Shifting Catamaran in Alaska

    25. Re:Build the datacenter in alaska by Anonymous Coward · · Score: 1, Funny

      Run some massive fiber along the big pipeline.

      What, and risk a backhoe causing major ecological damage?

    26. Re:Build the datacenter in alaska by DemingBuiltMyHotRod · · Score: 1
      Only disadvantage is that you're out on the edge of the Internet

      Well that... and the fact that transportation costs to Alaska are quite high.

    27. Re:Build the datacenter in alaska by virtual_mps · · Score: 1
      Assuming you're buying in bulk (Meaning you need large storage capacity too) you can get the price to within a few percentage of buying electricity. The additional cost of running on generator power when needed will be offset by the fact that you don't need cooling. When the grid is up and running, you'll be pocketing that money.

      Right up to the point where you get the big honkin' EPA fine for running an inefficient, polluting generator. We have huge diesel capacity here, but we never run the things except in a (real, not convenience) emergency because you get fined for running an emergency generator more than N days/year. If you want to avoid that you need to run a clean generator, probably off of natural gas rather than diesel and it will cost you a lot more than grid power in any reasonable locality in a first-world country.
    28. Re:Build the datacenter in alaska by C0vardeAn0nim0 · · Score: 1

      ever heard about radiation and what it can do to computers ?

      heck, the space station runs on i386s and i486s because the bigger circuits are more resistant to the radiation effects, and even those old suckers need a lot of shielding.

      --
      What ? Me, worry ?
    29. Re:Build the datacenter in alaska by evilviper · · Score: 1
      High energy protons spilling through the earth's magnetic weak spots causing massive spikes in any significant electrical circuits....

      Easily solved by just a few dollars worth of conductive material (ie. a faraday cage) installed into the building. Preferably hidden.
      --
      Slashdot gets worse every day... Pipedot: News for nerds, without the corporate slant
    30. Re:Build the datacenter in alaska by evilviper · · Score: 1
      Without the a/c on, my basement hovers around 68 degrees throughout the year.

      Being a few feet underground is slightly cooler than ambient air tempuratures at ground level. However, it's incredibly easy to keep any room cool when you DON'T have several kWatts of electrical equipment running constantly.

      Once you've got a lot of heat pumping into that room, the tiny tempurature advantage becomes insignificant.

      It's much easier to control the temperature than elsewhere in the house.

      Of course it is... That's where all the cold air is going, making the rest of your house warmer. Your house is also not a heavily insulated industrial building, which makes a difference.
      --
      Slashdot gets worse every day... Pipedot: News for nerds, without the corporate slant
    31. Re:Build the datacenter in alaska by LWATCDR · · Score: 1

      Not only all that but Alaska does not have the best connectivity to the Internet.
      And you would have issues with building on permafrost, transportation, and goodness knows what else.
      It could work if you wanted to use natural gas for power. As you pointed out Alaska has a lot of that but no good way to get it to market. And for cooling you could always tap a river. I do wonder if parts of Alaska and Canada might not be good places for data centers. I guess it depends on the cost of energy and the cost of building the pipes.
      Of course you have to love the idea of giant domed citys housing data centers.

      --
      See my blog http://ilovecookes.blogspot.com/ for light hearted technical information.
    32. Re:Build the datacenter in alaska by Anonymous Coward · · Score: 0

      Yeah, now it is pretty much perfe&^@£.NO CARRIER

    33. Re:Build the datacenter in alaska by OhPlz · · Score: 1

      Ahh, is that what happened? There was a trend about six years ago here in the Northeast where retailers and supermarkets were balking at the obscene electric rates we were paying. They went off grid and ran their generators 24/7. It sounded like it was going to be the new "big thing" and then it just simply died off. I had wondered what happened.

    34. Re:Build the datacenter in alaska by BlogPope · · Score: 1
      Well that... and the fact that transportation costs to Alaska are quite high.

      There's also the problem with the D/C melting the permafrost and sinking

      --
      My other car is a Popemobile
    35. Re:Build the datacenter in alaska by Anonymous Coward · · Score: 0
      Any benefits might be offset by having to have huge UPS and generator systems.

      Uh, any respectable colo will have those - along with stockpiling enough fuel to run those generators. Even in downtown san francisco high end colocation centers have to have those abilities.

      $64, 10.7GB disc, 264GB, $0 domain, shell, MySQL, cron, RoR, cvs. [dreamhost.com]

      Uh, for $49 I can get 80GB on a dedicated server from various places -- with the UPS & power capabilities I described. (though I admit it's awesome you guys have RoR pre-instaleld)

    36. Re:Build the datacenter in alaska by Anonymous Coward · · Score: 0

      Actually at some not-very-large scale, it's even cheaper. Quite a few universities run their own co-gen plants.

    37. Re:Build the datacenter in alaska by That's+Unpossible! · · Score: 1

      That sure does sound ecologicaly friendly too! :)

      I would wager a bet that it is more ecologically friendly than how your local power company is generating power for your grid (usually by burning coal, unless you live in one of those areas smart enough to be using nuclear energy).

      --
      Ironically, the word ironically is often used incorrectly.
    38. Re:Build the datacenter in alaska by deathazre · · Score: 1

      That sounds like complete and utter bull to me. I've never heard of such a fine -- and I work on gensets for a caterpillar dealer.

      In fact, one job I worked on was a sawmill where they used an old 3406 (mechanical, ~15L, 350kW -- same base engine used in a lot of trucks) prime-rated genset for their sole source of power. No utility hookup whatsoever.

      If you're really concerned about emissions, get an electronically-controlled genset and an SCR system.

      And if you want to save more money, consider peak shaving (running the generator when power demand is high in order to get a lower rate from the utility company -- and no, you don't have to push power to the grid).

      --
      Karma: Negative (Mostly affected by dorm trolling)
    39. Re:Build the datacenter in alaska by virtual_mps · · Score: 1
      That sounds like complete and utter bull to me. I've never heard of such a fine -- and I work on gensets for a caterpillar dealer.

      Ah, the "I'm ignorant, so it must not exist" premise--what a classic.

      Try googling for keywords like "PTE" (potential to emit), "EPA", and "Title V". The regulations depend on your location, the size of the generator, the quantity of pollutants, etc. In some cases you're small enough to not have to worry, in some cases you'll need to do paperwork, and in some cases you'll need a permit. The "number of hours per year for an emergency generator" rule is covered by an EPA memo which permitted sites to use a small per-year runtime to calculate their PTE, making it easier to duck under the minimums--but in return you couldn't run your generator too much or you wouldn't be able to call it an emergency generator and you'd have to file more paperwork. Failing to file the paperwork makes you liable to a fine.
  12. Mirror by Leroy_Brown242 · · Score: 1

    If someone has a cache, I'll host.

    email me the goods.

    thumper **AT** alderflats.com

    1. Re:Mirror by Mechcozmo · · Score: 1

      Is this called, "Slashicide"?

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

      Wow! He really is going cheap if he's using one of those first-generation Bondi Blue A/C units.

      Didn't want spring for the current top-of-the-line white, or even a more recent titanium or graphite?

  13. Crime Lab? by Saeed+al-Sahaf · · Score: 1

    I'll tell you Jeb, if that's how the Nebraska State Patrol runs a datacenter, makes me wonder about their crime lab...

    --
    "Who are in control, they are not in control of anything - they don't even control themselves!" - Glen Beck
  14. Ouch. by d474 · · Score: 1, Funny

    I hate to burst your bubble, but you just got slashdotted.

    --
    Authority questions you. Return the favor.
  15. Amazing... by Anonymous Coward · · Score: 0

    You have realized that if you keep the cold air near the hot things that they cool down better!

    Yes, that, and a "REAL" data center get your crap on /.

  16. Mirror by synthparadox · · Score: 1, Redundant

    A mirror of the thumbnails is available here.

  17. What's that sound? by Ethon · · Score: 2, Funny

    Did anyone else just hear a popping sound?

  18. It's misleading by SamMichaels · · Score: 4, Insightful

    1.5 ton sounds expensive, big and awe-inspiring. It's not. Most people have cheap 1 ton a/c units in their living room walls (12,000 btu). My 12,000 (1 ton) unit is barely able to cool 3 computers. Good luck with a datacenter.

    1. Re:It's misleading by TheFlyingGoat · · Score: 3, Interesting

      Your 3 computers must be built with awful components. At my former employer we were keeping the entire server room cool with a $200 window air conditioner. The server room had 2 HP servers, a handful of 1U IBM x336 servers, 3 workstations, 2 racks of networking equipment, and a variety of other smaller devices. The window unit ran about once an hour for 5 minutes in 85 degree weather and kept the room sufficiently cool.

      The key is to move your UPS out of the temperature controlled area, since that's the biggest source of heat in most server rooms. Then we were gradually replacing the multitude of old HP servers with the IBM servers. Each IBM could to the work of 4 HP's easily, and ran much cooler than a single HP.

      Your 3 computers must be running 1000+ watt power supplies, multiple 6800 vid cards, and a ton of hard drives if your home a/c is having trouble cooling that room.

      --
      You have enemies? Good. That means you've stood up for something, sometime in your life. --Winston Churchill
    2. Re:It's misleading by slashdot.org · · Score: 1

      Your 3 computers must be running 1000+ watt power supplies, multiple 6800 vid cards, and a ton of hard drives if your home a/c is having trouble cooling that room.

      Or maybe he just lives here

    3. Re:It's misleading by drmerope · · Score: 4, Informative

      No offense, but maybe you missed the point?

      The article was entitled "Keeping a Data Center Cool _on the Cheap_" (emphasis added)

      The suggestion being that they were able to significantly cut back on the size and power consumption of the cooling plant by using plastic wrap.

    4. Re:It's misleading by msblack · · Score: 1

      One and a half tons of cooling capacity is rather paltry. You need around 3 tons to cool a typical house. One ton will cool a 600 square-foot room, maybe a little more if well insulated.

      --
      signature pending slashdot approval
    5. Re:It's misleading by emerrill · · Score: 1

      Either your A/C unit is horribly bad (running below its rating) or you have some really beefy machines.

      I have a room with 2x(2x2.5GHz G5's), a dual xServer, and 3 Pent 4 machines, overall, with the UPS and various other equip, I draw about 1600W under normal load, and my ~.4 ton window unit keeps the room at a comfortable 65.

    6. Re:It's misleading by renehollan · · Score: 1
      Indeed. When I lived in Texas, in a 3200 square foot house, I had two zones of A/C totalling 7 ton: 5 ton for the bottom floor, and 2 ton for the smaller top floor.

      My electricity bill in July was around $350 extra just for the A/C to keep the house at a cool 80 degrees Farenheit (daytime outside temperature easily reached 100F and sometimes 110F).

      --
      You could've hired me.
    7. Re:It's misleading by Tim+C · · Score: 1

      I suspect you missed the point - I have zero knowledge of sizes/scales as far as air conditioning units go, and on reading the summary my immediate thought was "3 tons of AC unit doesn't *sound* cheap..."

      I suspect that that's what the OP meant - that 3 tons is actually very little, despite sounding (to the uninitiated) like an awful lot.

    8. Re:It's misleading by BigPappa · · Score: 1

      The key is to move your UPS out of the temperature controlled area, since that's the biggest source of heat in most server rooms.


      Problem with that is your going to have to swap out your batteries in the UPS something like every 6 months. The leading causes of battery failure is change of temperature and sustained above room temperature. We've had to replace at least half the wiring closet UPS batteries because they were getting in to the 80's or 90's in temp for a sustained period. That's a lot of money right there. We keep the UPS's in the room so the batteries will last for more than a year.
    9. Re:It's misleading by Braf · · Score: 1
      The key is to move your UPS out of the temperature controlled area, since that's the biggest source of heat in most server rooms.

      And the next step is to move the servers out of temperature controlled area as that's the next biggest source of heat in most server rooms.

    10. Re:It's misleading by pHDNgell · · Score: 1

      What are these three computers?

      I don't know tons, but the cheapest AC Fry's had ($100 for something like 5000BTU) keeps my machine room fairly cool. I've got an Indy, O2, dual Ultra 2 (and a full D1000), G4 cube, and four or five random PCs. There's also a couple of managed switches and possibly more stuff (including UPS).

      --
      -- The world is watching America, and America is watching TV.
    11. Re:It's misleading by An+ominous+Cow+art · · Score: 1

      > Each IBM could to the work of 4 HP's easily, and ran much cooler than a single HP.

      And only use up 50% more letters.

    12. Re:It's misleading by dumbfounder · · Score: 1

      that's because you don't have them wrapped in plastic with 20 rolls of duct tape. DUH.

  19. The "Slashdot Effect" just... by Anonymous Coward · · Score: 0

    ..let the air out of your bubble.

    Better call in the portable cooling units A.S.A.P :D

  20. this doesn't work... by Anonymous Coward · · Score: 5, Funny

    keeping a datacenter cool? Thats nothing..

    try moving to india to try and get your job back from dell only to get dysintry and heat stroke, lose your wallet and end up working in low grade indian miget porno to get enough money to buy a can of coke, only to get typhode.

    Bubbles...pfff...

    Ps. I'm writing this from Iran, send help.

    1. Re:this doesn't work... by Joe+Jarvis · · Score: 1
      try moving to india to try and get your job back from dell only to get dysintry and heat stroke, lose your wallet and end up working in low grade indian miget porno to get enough money to buy a can of coke, only to get typhode.

      Not to worry. I hear there are over 10,000 Indian doctors working on a cure for spellingitis.

  21. lame by Anonymous Coward · · Score: 0

    how fucking stupid is this?

    someone has to say it.
    let me be the first

    1. Re:lame by Sjobeck · · Score: 0

      It is as unintelligent as only someone who works in government could conceive.

      Agreed.

      Tremendously so.

      Let me get this straight, we just pissed $200+B down to the toilet that is democracy in the middle east but we can not afford to buy an APCC NetShelter?

      I hate this kind of bullshit.

      This dope publicized this like it is fun or funny or something; he ought to be embarassed. And fired. And docked. And publically so. The rest of these stooges take note.

  22. Net Timeout Error... Cool! by Anonymous Coward · · Score: 0

    The operation timed out when attempting to contact www.jebba.net.

    At least your melted blob of a server is cold.

    -- The /.ers

  23. Actually... by WindBourne · · Score: 1

    I think that it is great that they are looking to lower costs. It strikes me as they are trying to be inovative. Considering the outlandous federal defict and even my state's defict (colorado), I think they could use more like them.

    --
    I prefer the "u" in honour as it seems to be missing these days.
  24. Now he's got two possible problems by Stephen+Samuel · · Score: 1
    I'm wondering if the poster's gonna be in boiling oil Monday morning because:
    1. it's probably against some security policy to publicize pictures of the data centre, and/or
    2. (s)He's managed to slashdot their server(s).
    --
    Free Software: Like love, it grows best when given away.
  25. Now that it's been posted to slashdot by Anonymous Coward · · Score: 0
    from the data center: *pop* *pop* *pop* *pop* *pop* *pop* *pop* *pop* *pop* *pop* *pop* *pop* *pop* *pop* *pop*

    Gotta love bubble wrap!

  26. TWO problems? by Saeed+al-Sahaf · · Score: 1
    (s)He's managed to slashdot their server(s).

    The first point is possible, but do you really think "jebba.net" is a "government agency"? It's either a shared server at some hosting company, or his own on a DSL.

    --
    "Who are in control, they are not in control of anything - they don't even control themselves!" - Glen Beck
    1. Re:TWO problems? by someonehasmyname · · Score: 1

      65.28.169.11 (www.jebba.net) is a RoadRunner IP Address.

      --
      Common sense is not so common.
  27. OSHA? by Saeed+al-Sahaf · · Score: 1
    I think that it is great that they are looking to lower costs.

    Yes, but plastic tarp and duct tape ain't the way. In fact it's probably against OSHA rules, for good reasons.

    --
    "Who are in control, they are not in control of anything - they don't even control themselves!" - Glen Beck
    1. Re:OSHA? by wik · · Score: 1

      Didn't they suggest this combination when the "terrorists" were going to send chemical and bio WMD to our houses? Maybe they're just re-using useless surplus anti-terror materials.

      --
      / \
      \ / ASCII ribbon campaign for peace
      x
      / \
  28. Better yet by Mr_Icon · · Score: 3, Funny

    You want cool and cheap? Move your datacenter to the North of Siberia.

    --
    If you open yourself to the foo, You and foo become one.
    1. Re:Better yet by Anonymous Coward · · Score: 0

      LOL! =D

    2. Re:Better yet by NaruVonWilkins · · Score: 1

      This may actually be a more realistic idea than you think. Some Canadian universities use cold temperatures outside to cut drastically on their electric bill in winter already.

    3. Re:Better yet by tigga · · Score: 1
      You want cool and cheap? Move your datacenter to the North of Siberia.


      Yeah, if Internet connection's not required...

    4. Re:Better yet by lincarnate08 · · Score: 1

      Cool, yes. Cheap, I doubt it. If you moved your datacenter to the Arctic Ocean you'd have to buy new servers.

    5. Re:Better yet by Council · · Score: 1

      It sucks how the speed of light limit is such an absolute no-way-around-it problem, for a lot of applications. Guarenteed 10 ms round-trip for every thousand miles (5ms/1000 miles). So halfway around the earth and you've got 100ms ping guarenteed lower limit, even if you run the cables as straight as you can (on the surface, of course).

      Mostly I'm mentioning this because it was an excuse to put together a rule of thumb for light/distance in real-world estimation, like the 1-foot-per-nanosecond rule. 5.5 ms to go 1000 miles, rounded down and you've got a nice absolute floor.

      5ms per 1000 miles, 10ms ping. More because your cables aren't straight enough.

      --
      xkcd.com - a webcomic of mathematics, love, and language.
    6. Re:Better yet by iceborer · · Score: 1

      In Soviet Russia, your datacenter cools you...um, on the cheap...oh Hell.

    7. Re:Better yet by Not_Wiggins · · Score: 1

      Or here's a thought: why not locate datacenters in sub-basements? They're more secure from break-ins as there's only one route of attack (through the "roof" of the datacenter and not so much from the surrounding walls), and it should be easy to utilize geo-thermal heat dissipation; at least, it should make it easier to cool with conventional cooling if one could manage to build a heat-exchanger directly into the walls.

      --
      Diplomacy is the art of saying, "Nice doggie!" until you can find a rock.
  29. yes. by testednegative · · Score: 2, Informative

    its been said before, but for the sake of trying.

    PLEASE POST PIC/MOVIE THREADS WITH CORAL MIRRORS OR MIRRORDOT MIRRORS OR A MIRROR OF A KIND.

    partial thumbnail pics only mirror here:
    http://mirrordot.org/stories/e90ad5cab7cfb4869cc0f 7bbfed3ed64/index.html

  30. Re:Damn you Slashdot by darksider415 · · Score: 0

    Well, looks like the slashdot effect at work. However, in a few hours, it'll be up for the rest of the world to read.

    --
    And they wonder why I left Windows.....
  31. Cereal & Beer Works! by __aaclcg7560 · · Score: 1

    Wouldn't breakfast cereal and beer keep the data center cool? I mean the machines, not the admins. ;)

    1. Re:Cereal & Beer Works! by name773 · · Score: 1

      heh, great reference :) (his art is simply astounding)

      just as a tip, it might discourage people from being insulting if you handled their banter more gracefully. i'm sure you can think of something less confrontational than calling them retards ;) childish is more accurate anyway. but try to get away from that whole idea, maybe more towards something about your workout program... that has the effect of showing people that you're trying to cut down on weight, that you're strong, and both without being demeaning. maybe even work in a nerdy reference; i know you can come up with a good way to handle this.

    2. Re:Cereal & Beer Works! by __aaclcg7560 · · Score: 1

      I tried a variety of techniques to handle the "retards". Posting the F.A.Q. seems to work pretty good since I haven't heard from them since then. I have been getting a ton of virus-infected email on both my personal and website email addresses over the last few days that may be related. I'll keep what you said in mind when I eventually update the F.A.Q. again. Thanks!

    3. Re:Cereal & Beer Works! by name773 · · Score: 1

      no problem, thank you for considering it :)

  32. Others... by WindBourne · · Score: 1

    use the great lakes.

    --
    I prefer the "u" in honour as it seems to be missing these days.
  33. If that is a "real" data centre... by Anonymous Coward · · Score: 0

    I will eat my hat.

    If we wrapped anything up in bubble wrap - or did such a f'ugly work of that cabling, then I would worry less about saving on heating costs and put my money into buying techs who arent clowns.
    Real data centres have diesel generators, multiple fans on racks and even spacing of servers. As well as adequate cabling that is tidy and colour coded.
    Real data centres don't skimp on such things.

    Not to mention it adds time onto getting access to racks when you require swapping those hot swappable drives and PSU units... It would be just like christmas day every time you go to open up the racks to work out what the problem is!

  34. You should be using Pentium M - based computers... by Harry+Balls · · Score: 2, Interesting

    ...like we do.

    25 watts per CPU, 50 watts per system.
    Period.

  35. Well, look who it is... by Anonymous Coward · · Score: 0

    Nebraska State Cops...

  36. Saw this around 1988 by xs400 · · Score: 1

    Remember those XTs (640 KB) 10 MB hard-drive, 4.77 Mhz. The management did not want to a/c the office(India), so there were vendors who sold glass racks with a window a/c unit to keep the unit cool. And careful with the turbo switch...

  37. i can see the call from ready.gov on monday by atarione · · Score: 2, Funny

    hey have you guys seen our ductape and plastic sheets?

    --
    actually I am happy to see you, however that is in fact a banana in my pocket.
  38. 1.5 Ton? by Acidangl · · Score: 1

    I wouldn't feel comfterable using a 1.5 Ton to cool the servers i have at my house.

    We have 25 tons of cooling in our core server room (at a small community college) and are looking at replaceing our current solution.....or maybe i should ugilify my cabinets.

    --
    I'm a cucumber
  39. Heh, great minds think alike! by pla · · Score: 4, Interesting

    Having recently faced a similar problem (though on a much smaller scale), we came up with almost the same solution.

    As one suggestion, though, cardboard (in 4x8ft sheets) proved a lot easier to work with than plastic sheets. For starters, the plastic requires attachment at the ceiling, and will eventually come loose under its own weight; cardboard, with a single fold in the sheet, will stand upright and support its own weight for years, assuming not too high of a humidity level. For another, cardboard won't flap around and potentially block air intakes nearly so easily as plastic will.

    Believe it or not, though, what we found the most effective way to make use of barely adequate AC - Don't treat the room as a closed system. You've basically used the plastic sheets to build giant chimneys - Now take advantage of that fact, and along with a high volume fan above each rack, just exhaust the air at the top outside rather than recycle it back into the room... Think of it this way... You spec your cooling to work to perhaps 110F ambient, right? At the top of a full rack, with 50-60F going in the bottom, you probably have 120-130F going out the top. Does it take more work to cool 130F, or 110F, back to 50F? Not to mention, your normal ambient shouldn't come anywhere near 110F...

    1. Re:Heh, great minds think alike! by Anonymous Coward · · Score: 0

      Hmmmmm brilliant...

      First plastic wrap then cardboard in the DataCentre. I bet your insurers will be thrilled at all the nice flammables on your raised floor. Gotta love that cardboard dust potentially tripping your VESDA system too.

      Good work guys..keep it up...

    2. Re:Heh, great minds think alike! by Anonymous Coward · · Score: 0

      Gotta love that cardboard dust potentially tripping your VESDA system too.

      That's why you need to give it a coat of spray paint about once a week. (easily done without even power cycling!)

    3. Re:Heh, great minds think alike! by RandomJoe · · Score: 1

      Actually, depending on the relative humidity levels, it just may be easier (more efficient) to cool the 130 degree air. Dryer air takes less energy to cool. If the air has already been cooled to 50-60F, and hasn't had any humidity introduced (closed system), it will very frequently have less moisture content than the 110F air you are pulling from somewhere else.

      Now, if you put a humidity sensor in the return air (above the racks) as well as the replacement air (outside, whatever) whenever the enthalpy of the replacement air is lower, by all means use it and vent the return. (This is comparative enthalpy - based economizing, for HVAC systems.)

      Of course, the humidity sensors and controls will cost more than the plastic sheeting or cardboard...

  40. Re:Damn you Slashdot by Ava3ar · · Score: 0, Offtopic

    a site can be /.'d before a single comment due to the subscribers, damn you paying people

    --
    ¦^)= The Vengance Will Come =(^¦
  41. partial mirror by Leroy_Brown242 · · Score: 1

    I'll add more pics, if I get them.

    http://forumpics.info/cooling/


    1. Re:partial mirror by Anonymous Coward · · Score: 0

      A few thumbnails and a lot of ads. Nice.

  42. STILL down. by parasonic · · Score: 0

    Some random 12 year old with a web page could have posted the address under his "cool links" (yahoo, nickelodeon, etc.) and it would have gone down after the first click. Shazbot!

  43. Famous last words by Anonymous Coward · · Score: 1, Funny
    Here's some pics to check it out.


    Famous last words...
  44. Re:Damn you Slashdot by Adrilla · · Score: 1

    That's why it pays to have that little asterisk next to my UID. I get to read the stories before the servers explode, killing thousands upon thousands of innocent victims and tramples their cities and shoots lasers at giant moths....no wait...wait, no that's Godzilla. But still the asterisk, early story reading thing, that's true, I'm almost sure of it.

    --

    "Plans are for fools! Oglethorpe, the plutonian (Aqua Teen Hunger Force)
  45. Re:You should be using Pentium M - based computers by muhgcee · · Score: 0

    Hello Mr. Balls -

    I am impressed by the products you offer on your website, but I find the website a bit lacking in info. I am up all night . . . and if I can just get a few questions answered I'd probably order up one of those servers tonight. I sent an email to sales@hostingator.com so I figured I'd put a note here to try to "speed up" a reply from that address.

    Thanks.
    stu@fourmajor.com

  46. Note from the Entomology freak by morcheeba · · Score: 4, Funny

    "(more like a termite tent)"

    "Most termite species are tropical or subtropical, but a few live in temperate regions." I'd posit that even fewer live in a properly cooled data center. So, on the surface (no pun intended), this doesn't seem to be a good comparison.

    But reading further into the Wikipedia article: "Termites have biting mouthparts and are soft-bodied, of moderate to small size. They live in dark nests and tunnels, except when the winged alates emerge to leave their parent colony." When comparing termites to geeks, they both seem to have biting mouth parts and the geeks are definitely soft-bodied. And of smaller size. And, like the termites, true geeks live in their parent's basement.

    "Termites cannot themselves digest the wood that they consume." Few geeks can live on chewed-up pencils. So again, another similarity.

    Lastly, Termites construct extremely large and elaborate mounds to house their colonies. ... The internal structure of these mounds can be quite complex, with ventilation chimneys for active temperature control" Need I say more?!

    1. Re:Note from the Entomology freak by Anonymous Coward · · Score: 0

      Need I say more?!

      No. In fact don't say anything. At all. Ever again. Period.

    2. Re:Note from the Entomology freak by morcheeba · · Score: 1

      I think that's a compliment,right? I'll take it as one :-)

    3. Re:Note from the Entomology freak by LaCosaNostradamus · · Score: 1

      You have out-done yourself, going above and beyond the call of /. duty. Consider yourself No Prized. {clap clap clap}

      --
      [You have a stable society when some nut guns down a schoolyard and the law doesn't change.]
    4. Re:Note from the Entomology freak by Anonymous Coward · · Score: 0

      "Termites have biting mouthparts and are soft-bodied, of moderate to small size. They live in dark nests and tunnels, except when the winged alates emerge to leave their parent colony."

      Edwin was saying that they were members of the Hymenoptera family and self replying, "Well, well. Quite the nibs, eh?"

      "They are characterized by unusual distinctions of three regions of the body-head, thorax, and abdomen-and by the stack or petiole of the abdomen moves very freely on the trunk of the thorax."

      "You wouldn't fool me?

      "The female, after laying her eggs, feeds the larvae with food regurgitated from her stomach."

      "Try to keep it clean, my lad."

      "Both males and females are winged."

      "And why not?"

      "But the female pulls off its wings and runs about without them."

      "I question that. I doubt if even an ant would be such an ass."

      "It's quite true. It says so in the book. Have you ever seen ants fight?"

      "Not that I remember."

      "They rise on their hind legs and curve the abdomen."

      And, to my consternation and chagrin, whether because it was his intention to illustrate or because he found his squatting position cramping to the limbs, this was just what he did himself. He rose on his hind legs, and stood facing me, curving the abdomen-at the exact moment when I perceived Florence emerging from the house and walking briskly in our direction.

      It was a crisis at which a less resourceful man might have supposed that all was lost. But the Woosters are quick thinkers.

      "Hullo!" I said.

      "What's the matter?"

      "Have you dropped a sixpence?"

      "No."

      "Somebody has. Look."

      "Where?"

      "Under that bush," I said, and pointed to a shrub of sorts on the edge of the drive.

      As you probably conjucture, in saying this I was descending to subterfuge, and anybody knowing Bertram Wooster and his rigid principles might have supposed that such wilful tampering with the truth would have caused the blush of shame to mantle his cheeck. Not so, however. If there was a flush to be noted, it was a flush of excitement and triumph.

      For my subtle appeal to the young blister's cupidity had not failed to achieve its end. Already, he was down on all fours, and if I had posed him with my own hands I could not have obtained better results. His bulging shorts seemed to smile up at me in a sort of inviting, welcoming way.

      As Jeeves had rightly said, there is a tide in the affairs of men which, taken by the flood, leads on to fortune. I drew back the leg, and let him have it just where the pants were tightest.

      It was a superb effort. Considering that I hadn't kicked anyone since the distant days of school, you mighthave thought that the machinery would have got rusty. But no. All the old skill lingered. My timing was perfect, and so was my follow through. He disappeared into the bush, travelling as if out of a gun.

  47. That is NOT a data center. by supabeast! · · Score: 1

    If you can fit your computers into plastic bubble made of cheap sheeting and duct tape, that is NOT a data center. It would have been easier to do what most small IT shops do - stick the servers in a small office with one of those cheesy little 1.5 ton units and shut the door.

    Post again to /. when you're working in a real datacenter.

    1. Re:That is NOT a data center. by Anonymous Coward · · Score: 0

      Post again to /. when you're not a pretentious...

      Actually, don't post this crap to Slashdot ever again.

    2. Re:That is NOT a data center. by Anonymous Coward · · Score: 0

      Agreed. Stories like this should NEVER be posted to Slashdot. They're a waste of time.

    3. Re:That is NOT a data center. by Anonymous Coward · · Score: 0

      rtfa (read the f***ing article) moron
      they are cutting back
      as in they are giving up
      real data centers
      and using this silly
      idea instead.
      did I mention MORON

  48. Great physical security? by wyoung76 · · Score: 1

    Don't you mean great security by obscurity?

    1. Re:Great physical security? by Anonymous Coward · · Score: 0
      Not at all.

      Alaska has lots of cheep unforrested land (tundra). You can have a 1-mile perimiter around your data center that can enable you to easily stop any unauthorized vechicle (potential car bomb) that may approach.

  49. Easy... by polyp2000 · · Score: 1

    Move it away from india , back home.

    --
    Electronic Music Made Using Linux http://soundcloud.com/polyp
  50. I'd fire anyone... by Anonymous Coward · · Score: 0

    Who said they'd fire anyone based upon what units they use instead of their efficacy.

    Units are a tool. It's most important that you are familiar and effective with your tools than whether your tool is also the favorite of someone else.

    1. Re:I'd fire anyone... by masklinn · · Score: 2, Insightful

      Wrong, units are not mere tools, they're communication tools, and the most important is not to be familiar with the communication tools it's that everyone is able to use&understand them. The efficiency of a communication tool comes from the quality and reliability of the communication it provides.

      metric is the standardised universal way to communicate, hence the tool to use.

      Just as english is currently the standardized and *mostly* universal communication language and should therefore be used whenever avaible.

      --
      "The way we can tell it's C# instead of Haskell is because it's nine lines instead of two." -- wadler
    2. Re:I'd fire anyone... by brainburger · · Score: 0, Flamebait
      Just as english is currently the standardized and *mostly* universal communication language and should therefore be used whenever avaible.
      So are you saying two Welsh people at home in private should speak English, or that an Inuit family should simply dispense with Inuktitut?
    3. Re:I'd fire anyone... by Anonymous Coward · · Score: 0

      No, he's saying you wouldn't hire a Chinese person who only speaks mandarin as a tech support rep for your english speaking clients.

    4. Re:I'd fire anyone... by sowellfan · · Score: 1

      I agree that the metric system is better, but when you're dealing with HVAC systems in the U.S., trying to do everything in SI and then convert to English at different points would add a great deal of trouble. The programs that calculate cooling loads have English unit outputs, and expect English unit inputs. The catalogs for U.S.-centric equipment are figured in English units (there are SI catalogs out there, but they're hard to get, and the equipment itself is typically designed in multiples of tons or Btu/hr). So I'd support a conversion to metric, but it hasn't happened yet, so I'll stick with English for now.

    5. Re:I'd fire anyone... by Anonymous Coward · · Score: 0

      "but when you're dealing with HVAC systems in the U.S., trying to do everything in SI and then convert to English at different points would add a great deal of trouble" ...and so, boys, is how we get even one more reason to press in the States for changing to the SI as everybody and his nephew has already done everywhere else.

  51. Dangerous fire hazard by hherb · · Score: 1

    This is an outrightly dangerous fire hazard. I would strongly advise against it.
    It also traps moisture which probably will condense along the plastic and end up where you don't want it.

  52. -Government- Datacenter... by Scott+Swezey · · Score: 2, Interesting

    I don't get all of the attacks about this not being a "real datacenter." Sure, I wouldn't want any of my things run from his DC, but he did say that it was a government datacenter on a budget... Surely all of us have dealt with cheap government / school people at some time who refuse to put money where its needed.

    --
    Scott Swezey
  53. Re:this doesn't work... [typo error] by Anonymous Coward · · Score: 0

    Typhode is wrong, should be "typhoid".

  54. Thats a wrap! by Doctrinal+Enforcer · · Score: 2, Funny

    Big bubbles no troubles.

    --
    VERITAS VOS LIBERABIT
  55. This is not about AC... by KlaymenDK · · Score: 2, Funny

    ...it's about load testing his data center!

    "Here's some pics to check it out." ?? Dead giveaway!

  56. Re:You should be using Pentium M - based computers by adrianmonk · · Score: 1
    25 watts per CPU, 50 watts per system. Period.

    Oh, you want to play that game, do you? The Mac mini I'm using to write this message can do a tad better than the Pentium M system you describe. I presently have it hooked to my Kill-a-Watt(tm) meter, and the whole system is drawing 20W from the A/C outlet. That's with disk spinning and CPU mostly idle. If I peg the CPU by doing
    dd if=/dev/zero of=/dev/null, the power usage jumps to a whopping 30W.

  57. Notes from the Etymology freak by Anonymous Coward · · Score: 0



    The word 'bubble' seems to have developed from the verb 'burble'. Such use is extant in Middle English by 1300 AD.

    [e.g., 1303 R. BRUNNE Handl. Synne 10207 As {th}o{ygh} here y{ygh}en shulde burble out.]

    A variant form of the verb, 'bubble', soon appears:

    [1477 NORTON Ord. Alch. in Ashm. iv. (1652) 47 Remember that Water will buble and boyle.]

    This verb was nouned at around the same time:

    [1481 CAXTON Myrr. II. xxi. 113 The water of those wellis sprynge vp with grete bobles.]

    [1605 SHAKES. Macb. I. iii. 722 The Earth hath bubbles, as the Water ha's. ]

    Other Teutonic languages exhibit a parallel development of similar words.

    New words and phrases continue to be developed from 'bubble'. Hence,

    Bubble Chamber, Bubble Memory, Bubblegum, and (of course) everybody's favorite, Bubblegum Pop

    References:

    Almost all my information is derived from the OED entry 'bubble'.

  58. what about ducts? by adrianmonk · · Score: 2, Interesting

    I've thought about alternate methods for keeping computers cool, and I started to wonder about just feeding cold air directly into the intake of the computer itself, rather than trying to surround the whole computer with cold air. Then the computer's hot air output is not polluting your cold air with hot.

    What I had in mind is a sort of a system that would supply cold air through ducts (similar to the tubes that are used for hot air exhaust on a clothes dryer) at positive pressure. It'd then be a matter of just hooking these up to your fan intakes on the computer, and you'd have very cold air flowing straight through the system.

    One could easily supply the required cold air through ducts by putting a big cardboard box (or wooden box, etc., etc.) on the front of a window unit, then cutting holes and attaching hoses where required.

    I've wondered if anyone has tried something like this. The disadvantage is that you have to run new ducts every time you install a piece of new equipment. The advantage is that the computers are being fed with cold air directly after it passes through the air conditioner's evaporator coil while it's still cold, instead of reaching the computers after it has had a chance to mix with hot air in the room. Kind of like standing right under the A/C vent when you go indoors on a really hot summer day.

    1. Re:what about ducts? by aXis100 · · Score: 1

      If you use a rack properly (fill it and seal it), then the fronts (or bottoms) can be fed with cool air and the back (or top) exhausted. I have some customers that specially modified their racks to achieve this.

    2. Re:what about ducts? by Anonymous Coward · · Score: 0

      In the old mainframe days the computer center usually had a modular rised floor that held the cabling but also was kept at overpressure by the AC.

      Under big cabinets the tiles were left open and the cold air rushed up through the cabinets and out the top. The top usually consisted of a matrix of fans.

      For smaller equipment there were perforated tiles that let some amount of cool air stream through.

    3. Re:what about ducts? by tomlouie · · Score: 1

      That's not a bad idea. The only problem is that typical window air conditions only put out chilled air in bursts, during their duty cycles. The rest of the time, they just recirculate room air (to allow it refridgerant time to recompress, I believe). So during the recirculate part of the cycle, your PC would get room temperature air. Depending on your system, that might be ok, or disasterous.

      I've heard of people who pump air from cool parts of their house into their PCs. Here's one link:

      http://overclockers.com/tips1121/

      Tom

    4. Re:what about ducts? by suitepotato · · Score: 1

      Normal rack servers aren't really well designed for cooling no matter what they like to say, and are more like a self-heating pizza in a deliveryman's insulated carry case.

      Telecom style systems where the boards are in vertical slots with fans above, below, or both run a lot cooler. Put an AC type of cooler in place of the fans and you're there.

      However, you still need someplace to get rid of the heat. First from the equipment and second from the air conditioner. Best is to put the air conditioning outside with careful landscaping, maybe a manmade pond to throw a set of coils into, and pump cold compressed air into the building and straight to the equipment. Then its a matter of maintaining flow through.

      --
      If my grammar and spelling are off, I am [distracted/tired/careless] (take your pick)
  59. Forget the servers by Ours · · Score: 1

    Man, with the temperature rising in my office, I'll stick to trying to keep myself cool.

    --
    "You superiour intellect is no match for our puny weapons" - The Simpsons
  60. It's a green house by johansalk · · Score: 1

    What you effectively did is made a mini-greenhouse.

  61. That's not nice by lheal · · Score: 1
    So, we decided to wrap the whole thing in plastic (including two 1.5 ton ACs )

    Those admins look big, but nowhere near 3000 pounds each. Besides, if they would just log in they wouldn't have to post as AC any more.

    --
    Raise your children as if you were teaching them to raise your grandchildren, because you are.
  62. Cheap = Low quality by Minkey+Brines · · Score: 0

    In I.T., cheap doesn't mean low cost, it means low quality. Just because your Linux distribution says i386 doesn't mean that the spare 386 desktop should be used as a server... Likewise, any organization that depends on data handling infrastructure deserves what it gets when it tries to do things "on the cheap."

    1. Re:Cheap = Low quality by Anonymous Coward · · Score: 0

      By your logic, open source is the worst quality software in the world. (Granted, true for most of the work on Freshmeat or SourceForge, but there are exceptions)

    2. Re:Cheap = Low quality by Minkey+Brines · · Score: 1

      It is my experience that businesses short change themselves right out of quality. If one way to do that is to use Open Source or Free Software without paying someone for support (if you can't fully support it yourself), then yes that means low quality. There's no reason a company can't spend the same amount of money originally budgeted for closed source license costs on Open Source or Free software support. In many cases, I would say that these companies would benefit *much* more than those committed to closed source alternatives.

      Having said that, this is a much wider issue than just software. It boils down to how companies can't do without at least the appearance of quality (customers wouldn't stand for it) but that also won't pay for real quality (they treat it as if it just costs too much).

      A similar idea is the true cost of the beef you eat. What is the total impact to the planet for the production of the meat you eat?

  63. 1.5 ton Anonymous Cowards? by allanj · · Score: 2, Funny

    "two 1.5 ton AC"

    I know that lots of geeks tend to gain weight, but those are big Anonymous Cowards!

    --
    Black holes are where God divided by zero
    1. Re:1.5 ton Anonymous Cowards? by Anonymous Coward · · Score: 0

      They are not anonymous cowards, they are gianormus cowards!

  64. one small mistake: by hummassa · · Score: 2, Interesting

    Kelvin is capitalized to differentiate it from kilo.

    --
    It's better to be the foot on the boot than the face on the pavement. ~~ tkx Kadin2048
    1. Re:one small mistake: by Guppy06 · · Score: 2, Informative

      The abbreviations of units based on proper names (K, J, W, N) are capitalized, but when spelled out they start with a lower-case letter (kelvin, joule, watt, newton).

    2. Re:one small mistake: by Anonymous Coward · · Score: 0

      Yep... my favorite example of this is the decibel (dB), named for Alexander Graham Bell.

      Well, really the decibel is a unit derived from the bel, but I've never seen anyone actually using the unit 'bel'.

    3. Re:one small mistake: by wizardhat · · Score: 2, Funny

      Look at any Seagate product specification page.

    4. Re:one small mistake: by AyeRoxor! · · Score: 1

      Kelvin is capitalized to differentiate it from kilo.

      While Kelvin is represented as a capital K, there is no such unit as the kilo, and no circumstance where a kilosomething and a Kelvin could be confused; can ya comprehend?

      /mea culpa. Alliteration keeps capturing my concentration...

    5. Re:one small mistake: by wgaryhas · · Score: 1

      technically it is acceptable to measure something in kilokelvins (kK), but I don't think it has ever been done.

      --
      "For every complex problem, there is a solution that is simple, neat, and wrong." - H.L. Mencken
  65. Re:Cooked Books by Doc+Ruby · · Score: 1, Insightful

    Moderation -1
    100% Flamebait

    This political TrollMod'ing to suppress criticism of Bush is really tiresome. Go ahead - flame me, I can take it. I'll just slice the flames to ribbons with the simple facts that elude the flamers. Their flames are their problem, anyway, if that's all they've got. FWIW, how come there's no "Flame" moderation?

    --

    --
    make install -not war

  66. 1.5 tonne AC... by howman · · Score: 1

    looks like two of them were not enough to keep thier servers going up in flame... "Pics to check out..."
    They either wanted to load test their systems or they forgot to read the warning on the bubble wrap that said don't inhale the air from inside the bubbles... My I suggest ten times as much cooling piped directly around your CPS's if your are going to even think about positng "Pics to check out" on /.

    --
    flinging poop since 1969
    1. Re:1.5 tonne AC... by Anonymous Coward · · Score: 0

      Ton, not tonne, limey bastard.

  67. What about us less fortunate? by grimJester · · Score: 1

    end up working in low grade indian miget porno to get enough money to buy a can of coke

    I'm not a midget, you insensitive clod!

    1. Re:What about us less fortunate? by Anonymous Coward · · Score: 0

      eh.. When you're desperate for work, you'll learn to fake it..

  68. Ventilation vs AC by spectrokid · · Score: 1

    Assuming you live in an area with outside temperatures below 40 C (104 F), is it not a lot more energy efficient to build strong ventilation with outside air rather then an AC?

    --

    10 ?"Hello World" life was simple then

  69. Re:You should be using Pentium M - based computers by Anonymous Coward · · Score: 0

    I'm typing this from an iBook but come now, that Pentium M at 25 watts is probably doing as much as a P4 at 50. Meanwhile your (and my) G4 just do not have the muscle to compete.

    If you don't need the added brawn, that's fine - if not you're in a world of hurt if you don't have it.

  70. Re:You should be using Pentium M - based computers by ErikZ · · Score: 1

    Heh. So, which has more CPU power? A G4 or a Pentium M?

    --
    Democrats or Republicans. They are both taking us to the same place and they are not afraid of us anymore.
  71. why not styrofoam by jksn · · Score: 1

    plastic is not good enough to keep the cold temperature inside. better using styrofoam layered with alloy panel as a wall

    1. Re:why not styrofoam by BK425 · · Score: 1

      One other possibility would be to use plasterboard fastened to structural members (sometimes they're called "2x4s" or "metal studs" that are surrounded by insulation. I think they call this ... walls.

  72. Why? by madshot · · Score: 1

    Why? oh wait.. maybe I should ask that question.. ugg.. geeks..

    --
    Obama = Socialism.
  73. Bublewrap is for popping by Anonymous Coward · · Score: 0

    Silly admins, bubble-wrap is for popping.

    I want savings proof. Please provide the list with a costs and savings breakdown over a period of time.

    And for the curious, amount of bubbles popped over a period of time might not be bad also.

  74. Is this actually saving energy? by shoemakc · · Score: 1

    Has there actually been any sort of study to prove this new setup is more effcient? Aside from the reasons mentioned above:

    -High temperature differentials on the insulation
    -Lost productivity from datacenter workers

    How about the fact that by undersizing the a/c you may be forcing the unit to act very ineffciently? I'm no mechanical engineer, but i work with them and from what i understand either undersizing or oversizing an a/c can ruin the effciency of the whole system.

    Also you need to take into account that the heat needs to be dispersed twice, once from the rack to the space and again from the space to outside, compounding ineffciencies.

    I suspect this could be a potential "Penny Wise, Pound Foolish" sort of situation. You know, those same people who balk at spending $10,000 more in equipment to save $20,000 in energy over 3 years.

    -Chris

    --
    --an unbreakable toy is useful for breaking other toys--
  75. Emergency shutdown...Aiiee by mfarver · · Score: 1

    I have had the same type of problem.. too many servers, to little Air Conditioning. The accountants have no problems with buying 5 more servers to support a particular project, but spending $10k on AC upgrades is not permitted.

    In our case the AC could hold the room in the mid 70F running continously. This lead to a compressor failure every 6 months and no backup AC. Temps in the room would quickly soar in the 100-110F until servers began failing or shutting down. (This failure always occurred on weekends, when the building air was shutdown and the datacenter was cooling its normal load plus leakage from a 90F degree building thru uninsulated walls.) By the time IT staff made it onsite several servers would be dead.

    The solution was temp based emergency shutdown. Take a DB9 female connector and solder a 10k resistor between CTS and DTR and attach a two connector cable to RTS and CTS. Run the two wire cable to the Y and R(or RC) terminals of a common home thermostat. Set the thermostat to cool mode and the setpoint to 85. Plug the DB9 into a spare serial port.

    Install UPSd or use the UPS tab in windows to assign a generic UPS to that port and set a 5 minute shutdown on AC failure. When the thermostat turns on (indicating temp is above the 85F setpoint) the cable will simulate an power failure on a old dumb UPS and shutdown the system.

    1. Re:Emergency shutdown...Aiiee by BK425 · · Score: 1

      Or, make a nice spreadsheet of time costs (I'm hoping you were paid when called in to shovel melted cpu off the floor) and costs of replacement systems and have the system built properly.

      I'll let a skeleton out of the closet here (sheild your eyes) I know some accountants... the accountants I know like to account. They wouldn't hide from quantifying repair costs for systems that fail and down time costs for systems that are shut down (orderly or not) for lack of air conditioning.

      I think as engineering types we love to mcgyver things but there comes a time to step back, look at the bigger picture and -say- "this ain't makin a lot of sense here". Plastic sheets, cardboard, et al... it's real inventive I'm sure, but it isn't -walls-.

      They make walls out of plasterboard, structural material and insulation for a -reason-. Sure, they didn't cover those reasons in our CS courses but I bet there are reasons nonetheless...

  76. Well that it by EEPROMS · · Score: 1

    OK Im selling my stocks.. I just saw another "tech bubble" *ducks*

  77. Alaska's Real Estate on The Rise!!! by Klanglor · · Score: 1

    Alaska's Real Estate on The Rise!!! World Wide Interest in Outsourcing Datacenter to Alaska. The following is a Map of the newly build Megapolis in Alaska DDD DMD DDD D:Datacenter Facilities M:Megapolis Due to the high temperatature of Datacenter Heat from the various Datacenter surounding the Megapolis, Eng and Geeks working on the facility enjoy tropical weathers in the megapolis. Secure your piece of DigitalGold Now!!!!

  78. Hello, Mr. Fire Marshall. by mosel-saar-ruwer · · Score: 1

    As one suggestion, though, cardboard (in 4x8ft sheets) proved a lot easier to work with than plastic sheets.

    Have any of you guys ever heard of the concept of a "fire hazard"?

    Believe me, you do not want to be within a city block of this thing when it decides to go up in smoke.

    1. Re:Hello, Mr. Fire Marshall. by iggymanz · · Score: 2, Funny

      heh, the plastic folks maybe weren't acquainted with Mr. OSHA and Mr. Confined Space and Mr. Oxygen Deficency Hazard, so adding Mr. Fire Marshall to the list of ignored people isn't a big step.

    2. Re:Hello, Mr. Fire Marshall. by jedimaud · · Score: 1

      It's amazing how you can tell what kind of plastic it is just from some pictures.

    3. Re:Hello, Mr. Fire Marshall. by Anonymous Coward · · Score: 0

      Mod parent up. Funny or insightful, either one.

  79. Gvt. Agency Allows Pictures? by seigniory · · Score: 1

    How did the "government agency" allow the pictures to be taken? Even in my private datacenter, there's a strict no photography policy.

    Just curious.

    1. Re:Gvt. Agency Allows Pictures? by jedimaud · · Score: 1

      I took the pictures very carefully, and had to show a friend of mine. He thought I should slashdot it.

  80. Rack legs.... by ivan256 · · Score: 1

    I don't know if the submitter is reading the comments or not, but somebody should tell him...

    The rack legs aren't for earthquakes. In an earthquake the rack would just tip to the side anyway. The legs are there to prevent some moron who slides out all the rail mounted equipment in the rack at the same time from getting crushed by a falling rack.

    1. Re:Rack legs.... by CXI · · Score: 1

      You realize that you are trying to reason with a person that thinks it's a great idea to wrap their servers in plastic held up by a couple of pieces of duct tape as some kind of long term solution. They probably have all the servers in backwards because the legs kept getting in the way!

    2. Re:Rack legs.... by ivan256 · · Score: 1

      Hey, don't knock the plastic thing.

      You know those big plastic strips they hang in front of the doors in loading docks to keep the heat in? They're great if you share a room with your racks and want to keep the cooling for your equipment seperate from the rest of the room without building a wall. Why waste your cooling on the empty space, or refrigerate your office to keep your cluster cold? It does a great job of blocking the noise too. This guy had a decent idea... Except that I wouldn't use 4 mil plastic and duct tape. Get the good stuff. Or in his case I'd say "Do some research". If he asked around or searched with google he would have found a good way to do this and not have made a fool out of himself promoting his half-assed job on Slashdot as a novel creation.

    3. Re:Rack legs.... by Anonymous Coward · · Score: 0

      I'm guessing Henry Ford's first introduction of a creative idea met with the same type of critism. He eventually got it worked out. While the method here isn't the best or most sound... the solution (or rather the quest for one) is important.

  81. Datacenter heat biggest industry problem I've seen by br00tus · · Score: 1
    I've worked as a sysadmin for many years, and have seen data centers from New York City to San Francisco, from "tier one" co-location facilities to Fortune 100 server rooms. The one problem I've seen in virtually all of them: heat. The server rooms are too hot - if the servers are stacked on top of each other, especially in a cabinet, that makes them hotter, and if you consider what the temperature is on the surface of the CPU itself - that's hot. No matter who it is, or how much money the company has, it seems they all have heat problems, I only remember a few large data centers that had decent heat.

    While the norm in my view is "too hot", I've noticed that when things get too hot, which can mean in the high 20's Celsius (80's Fahrenheit), low 30's C. (90's F.), or even high 30's C. (100's C.), machines start breaking down. And usually they begin to break down one after another. I've worked at some Fortune 100 companies that could probably trim their sysadmin staff by a couple of people if they just kept their server rooms a little cooler. I've also had situations where when a room reached a certain temperature, quantity turned to quality and machines began breaking down one after another.

    In terms of me telling people whose rooms are too hot, I always do. Usually the smaller the company, which means the smaller the room, the more they listen to me. Big companies, big rooms usually just ignore my protests that 30 degrees C. for a machine is too hot.

  82. Hrumpfff by WillRobinson · · Score: 1

    Currently we have 12 servers, running inside an warehouse in Texas, with NO airconditioning. Proper ventlation of the servers and room, and we have experienced NO problems for over two years.

    Of course these are older servers, that have plenty of room and clearance in their cases. Those super compact R1 servers with a ton of heat in a small area are just asking for problems. All of that is due to centers thinking they need to have a 40LB cat in a 10LB bag.

  83. They forgot... by Mr.+Moose · · Score: 1

    ...to cut a hole in the plastic for the network cables, cause the server is not responding.

  84. two 1.5 ton ACs by Anonymous Coward · · Score: 0

    I've got a +9 oger slaying knife and I still can't hit those AC's!

  85. Its possible by Anonymous Coward · · Score: 0

    Before upgrading to our new datacenter, we used a 3 ton unit to cool our servers - and we had 5 racks of servers with 3 racks of networking equipment.
    Ok, it kept the room around 75 and we needed a fan at the back of the room to circulate the air, but it worked.

  86. Always wondered why racks don't account for this.. by WareW01f · · Score: 1

    Having spent many hours in server rooms, freezing my butt off, (As you usually only have to do major work in the summer, on weekends, when your dressed for it.... not) I've often wondered why racks/servers aren't designed with an intake on one end and an outtake on the other. Or at least just a sealed rack with a connection at the top/bottom. (It'd be like a freezer section for servers, you could even leave notes, like IPs in the frost on the glass!) Now granted it would most likely be doomed to be vendor specific, but we're used to that with the rails already, right? >;^)

    Seriously, you'd think cooking the units would be much more efficient. You could lock down things like humidity, and even cool things well below what makes sense now (power wise that is).

    Oh well, I suppose all of that makes way to much sense.

  87. Don't try this in YOUR datacenter. by deacon · · Score: 5, Funny
    Hey Kids!

    Ever wonder why computers are made of metal? Why the plastic pieces are made of plastics that meet flammability performance standards?

    Tiny Tim raises his hand

    Yes, Tiny Tim?

    Please, Mr Deacon Sir, so that if some source of ignition is present, the computer does not turn into a fireball spewing deadly poisionous smoke, Sir!

    Very Good, Tiny Tim. And what happens if someone hangs up huge sheets of generic, flammable plastic in an area with lots of potential ignition sources?

    Please, Mr Deacon Sir, sooner or later it catches fire, and people die. If these boneheads are lucky, Sir, someone from the Building Facilities or the Building Inspector will see this website and make them take it down, Sir.

    Very good, Tiny Tim. The rest of the class is to read up on Flammability.

  88. physics by arete · · Score: 1

    In the general case:
    A. It is much more efficient to cool/heat only exactly what you need cool/hot.

    B. It is usually more efficient to use one big cooler than many small ones - but it can depend massively on the design/cost/age of the coolers/heaters. Generally, every reversible process generates waste heat. So, unlike ACs, _typical_ electric heaters can be 100% efficient and a heatpump is technically more than 100% efficient at generating heat in your room (based on "thermal energy changed in your room / power energy used " )

    So if you wrap all the servers in a small space with the inside end of the ACs, you save a lot of power from "A". Since you only have accidental AC left for the rest of the room, it can get quite hot - but only if something else makes it hot (poorly insulated exterior walls, people, sun-facing windows...)

    --
    Looking for freelance Actionscript (Flash/Flex) or ColdFusion work and/or freelance developers. Email me, put Slashdot
  89. For a savings of.... by mr_z_beeblebrox · · Score: 1

    I work for a government agency that, like most, is trying to cut back some costs, and one of those costs is a REAL datacenter. So, we decided to wrap the whole thing in plastic

    A lot was saved by the government especially when they found that an existing vendor would sell them the bubble wrap at under 8 grand a square foot.

  90. Saving Money? In Government? Plueeeze... by Anonymous Coward · · Score: 0

    Any money saved is vacuumed off to nowhere; definitely not into the Systems/Sysadmins budget.

    Has anyone really noticed how much Government employees make? (no, not the obvious GS schedule for low level jobs).

  91. Obligatory Yakov Smirnoff by Anonymous Coward · · Score: 0

    In America you cool the data center.
    In soviet Russia central data puts you in the cooler.

  92. Fire hazard by jbrandv · · Score: 1

    I don't know the regulations where you are, but the fire department here would have this plastic removed after the first inspection. This doesn't meet fire code. If that bubble wrap/plactic ever cought fire you might die from the fumes before you could evacuate the building. The sys admins may even be liable for the deaths of their co-workers. If the management approved they too would be in court. Years in jail are too high a price to pay for some managers who are too cheap to buy proper HVAC.

  93. all the money he saved by confused+one · · Score: 1

    in cooling costs is now being used up in bandwidth expense...

  94. cool? its a bit amatuer. by museumpeace · · Score: 2, Informative

    Starting at HP labs, Ratnesh Sharma began work on the problem of cooling server farms two years ago.
    Then work with the university of Virginia evolved from that research. Finally, in work done with Duke U. it paid off in the form of software tools that were reported at Usenix'05 [you can ignore password pop-up if you go thru the google cache] as saving 25% of cooling costs, thats can be over $1000000/year for large data centers by dynamically distributing work load to machines that are running cooler by using temperature data as input to the load balancer. [if you can get at the usenix art., Duke has basically the same paper on line. Or just read the the Usenix abstract]

    --
    SLASHDOT: news for people who can't concentrate on work or have no life at all and got tired of yelling back at the TV.
    1. Re:cool? its a bit amatuer. by Anonymous Coward · · Score: 0

      And I suppose not using your spelling checker is MORE professional?

  95. "a couple locomotive engines" by da5idnetlimit.com · · Score: 1

    Last Data-Center I visited had two helicopter turbines connected to the generators outside.

    They have to test launch them every month or so, which is quite of a racket.

    Energy wise, they could run 2 months without electricity, had their own (largish) water tank and were last seen discussing putting a sterling engine somewhere in the AC system so as to win a little something back from all that heat.

    I met the "Support" team for the building...
    Well, they were geeks. No IT geeks, but the sort of people that do plumbing and electricity and networking in a very pro way.

    I saw both IT and support team working together when testing some Clustering tools.

    How do you test IRL maximum latency on a fiber network for a real-time clustering soft ?
    You take a Fenwick, two gigantic Fiber rolls (almost one mile fiber per roll), a "fiber crancker" and 3 servers connected to a nice and expensive fiber switch.

    Lol. Now that is IT 8)

    --
    It takes 40+ muscles to frown, but only four to extend your arm and bitchslap the motherfucker
  96. No way possible by Anonymous Coward · · Score: 0

    There's no way you could do this with bubble wrap.
    Everyone within 50 feet of the datacenter would be
    continually popping the bubbles. I mean that stuff
    is just irresistable. pop, pop, pop...

  97. April Fool's Already? by blooba · · Score: 1

    This has to be the dumbest idea I have ever seen on /.

    1. Re:April Fool's Already? by Anonymous Coward · · Score: 0

      Hmmm... what an ingenious way to get some cheap Accounting / Mgmt. types to fund a new datacentre. Mock up a "hey look how bad it is here" photoshoot, post it somewhere to get attention. Get a bunch of replies about OSHA violations, FireCode violations, potential deaths, etc. Get said Accounting / Mgmnt. types to click anonymously recvd. e-mail link to story.....

      These guys may be smarter than we are all giving them credit for... then again...?

    2. Re:April Fool's Already? by COMON$ · · Score: 1

      Here here, I think people are missing out on the beaurocracy behind gov't issues. It is one big card game.

      --
      CS: It is all sink or swim...oh and did I mention there are sharks in that water?
    3. Re:April Fool's Already? by jedimaud · · Score: 1

      No Comment. That would be a good idea....

  98. Uh.. unsafe as hell by EvilStein · · Score: 1

    I bet the fire department inspections will go over REALLY well.

    Gotta love putting up flammable materials that emit toxic chemicals when burned in a data center.

    Great idea, guys. I'm sure the insurance company will be thrilled when they find out.

    Sheesh, at least get a couple of those ice bucket homebrew air conditioner things posted a few weeks back. Not that putting buckets of ice water in the data center is a better idea, but they're far less likely to go up in flames. :P

    1. Re:Uh.. unsafe as hell by Anonymous Coward · · Score: 0

      Assuming it's all flammable, most drywall/insulation sheeting is non-flammable and non-toxic..... Think they were that smart?

  99. No, you want to be right or undersized by bluGill · · Score: 1

    A little oversized is okay for cooling, but you really want to have the right size, or not quite enough. A overpowered cooling until will cool your house down without removing the humidity. A underpowered unit won't get your house as cool as you want, but will be more comfortable because it removes all the humidity, allowing sweat to cool you down.

    Best if no AC as all. Unless you are sick or otherwise unhealthy, you don't need it. Drink a lot of water, and sit near a fan which uses less energy. (Yes this is what I do even though I have AC). Go swimming.

    1. Re:No, you want to be right or undersized by ckaminski · · Score: 1

      I use my AC on fan only, and it chills my 2nd floor bedroom right down. two window fans, both blow in, and my life is pleasant.

  100. Just get a powerfull fan by geekoid · · Score: 1

    and vent the heat out. perferably into the heating system of the building.

    --
    The Kruger Dunning explains most post on /. http://en.wikipedia.org/wiki/Dunning%E2%80%93Kruger_effect
    1. Re:Just get a powerfull fan by gedhrel · · Score: 1

      That isn't all it takes. AirCon is about the whole environment: that includes keeping humidity levels safe, too (not too high, not too low), and not just worrying about the temperature.

  101. ROFLMFAO by Anonymous Coward · · Score: 0

    That has to be the cheesiest "data center" I have ever seen! What are those, portable or window AC units? OMG! Calling a room full of computers a data center is a bit much. If you say you can't afford a real data center (raised floor, floor standing CRAC units, controlled access, conditioned power, UPS, generator, etc.), then you're risking your data and your business. I have an idea, try managed hosting or find a new career because when the fit hits the shan, you're gonna wish you were dead.

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

      "Cheesiest?" Ironic because I think we're being baited.... ooops, yep, I've replied to this trash now too... someone get the hook out of my mouth please?

      Maybe it's a bet to see if a certain number of posts can be generated... or maybe to see if they can get their 15 minutes like the SWK. It just smells...cheesey. You called it.

      I'd better get back to work or I'll end up working in a datacenter like this too.

  102. Re:Datacenter heat biggest industry problem I've s by NerveGas · · Score: 1


    The heat is actually the evil twin of the industry's other largest problem: Power delivery. As processers consume more heat, and customers install more and more CPUs, a LOT of data centers are struggling to deliver enough electricity, and then they're struggling to keep the place cool.

    It's the flip side of the same problem, but while you can easily sense that the room is getting really, really warm, you can't sense that their power feeds and UPS systems are very near capacity unless there's a failure.

    steve

    --
    Oh, you're not stuck, you're just unable to let go of the onion rings.
  103. Penny wise and pound foolish by Anonymous Coward · · Score: 0

    This is just plain dumb. What's the expected lifecycle of this ridiculous setup? How much trash and expense will you incur doing this, vs. creating and maintaining a proper facility? This is not ingenious or even practical. Our throw-away society is bad enough as it is, without encouraging people to create whole rooms which are basically garbage.

    1. Re:Penny wise and pound foolish by Anonymous Coward · · Score: 0

      Agreed but I'll assume the admins have no $$ approval.. Are they to blame or is the Organization's Administration? Any admin would love to have the proper facilities, I'm sure they would too. Suppose this datacenter would have already burnt up if they'd have done nothing?

  104. Doh! by Kent+Recal · · Score: 1

    (including two 1.5 ton ACs)

    Wow. I, too, post anonymously every now and then. But I'm not THAT fat.
    Respect bro'...

    1. Re:Doh! by Anonymous Coward · · Score: 0

      Hey, a data center condom....

      I bet you could hang meat in that bubble. Not sure who the bald goon is, but he sure is handsome.

  105. English Missile Bunkers work pretty well by billstewart · · Score: 1

    The Bunker is a data center housed in a former missile base in the UK. It's underground, which makes cooling much easier, and has highly redundant commercial power feeds in addition to the stuff they've added themselves. Being able to recycle that kind of space kept their construction costs low and keeps their cooling costs low, and it also looks cool, which has been useful for them to sell service to bankers and other reliability-paranoid types.

    --

    Bill Stewart
    New Fast-Compression-only CPR http://preview.tinyurl.com/dy575ks
  106. Fixed and Variable Costs and $500 hammers by billstewart · · Score: 1
    Back in the 80s, I worked for a defense contractor. There are lots of things that lead to the $500 hammer problem
    • Overspecification - The customer lists *all* the specs they like from *all* the products on the market, even though no single product meets all those specs.
    • Purchasing Bureaucrat requirements filtering and process formality - The vendor doesn't actually get to talk to the end user who wants the product - the vendor talks to a bunch of purchasing bureaucrats who only have a vague notion of what the end user wants, so they pad some of the requirements to be sure, and the end user has usually padded some of the requirements because he knows that the purchasing bureaucrats don't have a clue what they're doing. And the vendor can't simply sit down and talk to the end user about it, because purchasing rules usually prevent it (to prevent a whole 'nother set of ways that vendors can pad costs and rip off the government.) And usually some PHB in the process read some article in Jet Engine World or Hammer Manufacturer's Digest about some cool new innovation in jet engine hammers and wants to make sure that the end user's don't get ripped off by some vendor selling them an inferior hardware-store hammer.
    • Cost Accounting for Contract Issues. When you need to buy a $5m jet engine and a $5 jet engine hammer, it's going to cost at least $1000 for the paperwork, and that gets accounted for as part of the items being purchased. So you end up with two line items in the contract, a $5,000,500 jet engine and a $505 hammer.
    • Fixed vs. Variable Manufacturing Costs and Small Production Runs - Most manufacturing processes have a fixed cost to set up the manufacturing equipment and a variable cost that's the per-unit cost of making the items. So if it costs $5000 to set up the hammer-making equipment and $1 in variable costs for each hammer, Walmart is going to make a million hammers at a cost of $1+(5000/1000000) = $1.0005 each, mark them up to $10, and hold a big price-rollback sale with them marked down to $5 (which includes about a buck a hammer for advertising.) But the Air Force is going to order one mil-spec jet engine hammer with each of the ten jet engines they're buying, so they're going to pay $1 + (5000/10) = $501 per hammer, and then they're going to order a dozen mil-spec helicopter rotor hammers, which are slightly different from jet-engine hammers, so the machines need to be set up again, and they'll pay $1 + (5000/12) = $417.66 per hammer.
    Would it be cheaper to order a slightly *better* hammer that can be used for both, at a cost of $2 + (5000/22) = $229/hammer? Yup, but nobody could justify the cost of doing the paperwork to explain why they're buying fancier hammers, leave aside the costs of convening a MIL-SPEC-Joint-Tactical-Hammer Design Committee, or dealing with the auditors who have to explain why Halliburton is importing thousands of $100 MIL-SPEC-Joint-Tactical-Hammer units to Iraq where they're going to use them for busting up sheetrock (which could be done with $5 Walmart hammers) or using them in place of the $500 MIL-Spec-Titanium-Tank-Wheel-Hammer which is made by a contractor in Senator Foghorn's district.
    --

    Bill Stewart
    New Fast-Compression-only CPR http://preview.tinyurl.com/dy575ks
  107. Re:Always wondered why racks don't account for thi by trunksy · · Score: 1

    Um, that's not feasable for high-density environments such as "real" data centers because of the heat that's generated. When you're housing foods that don't generate large amounts of heat, the "freezer section" works. But when you start getting into stuff that actually creates heat, you have to have a medium to move the heat away from the source. That would be the general trend to move air from the front of the rack to the back of the rack. That's why rack's usually have about 60% of the front and back doors open for air.

    When you get into higher density computing centers, it becomes even more important to move air from front to back to be able to move enough heat away from the computers to keep them operating efficiently. Thus, the "freezer section" idea would not be possible. If you want to get into computational fluid dynamics, Liebert has equipment that can create "hot aisle/cold aisle" rack arrangements within a data center. Obviously, since you've never heard of Liebert, you probably work in a smaller, low-density data room. Liebert's cooling systems are used in medium to high density computing such as supercomputing applications.

    I'm assuming these guys are government workers for localities because I would be horrified if this were any permanent federal facility. I work for a federal facility and my current location's server room is somewhat "ghetto" too but only because we're a transient facility that will probably shut down in a month or two. Our regional facility, uses good ole' Liebert systems.

  108. Re:Datacenter heat biggest industry problem I've s by Anonymous Coward · · Score: 0
    if the servers are stacked on top of each other, especially in a cabinet, that makes them hotter


    That's a misconception that a lot of people have.. If you have 2 1U servers both with surface temparatures os 100deg putting them next to eachother is not going to make it more than 100 deg... it will still be 100 deg.. if you separate them and put say 1/2u of space between each server it won't make them any cooler, rather it will allow the cool air to leak to the back of the rack without doing the server any good.
  109. Thats pretty cool. Literally. by MrArmyAnt · · Score: 1

    I really Like this idea, may go run some ducting from my AC vent into the inlet on my PC in my room, I can sweat as much as neccessary, as long as my dang shuttle doesnt keep overheating. Case temp alone with a box fan can be up to 80 C.