Catch up on stories from the past week (and beyond) at the Slashdot story archive


Forgot your password?

Data Centers Breathe Easier With Less Oxygen 392

PC World is reporting that some companies are looking at a new method of fire protection in their server closets, oxygen-deprivation systems.""Wood stops burning when the oxygen content falls to 17 percent and plastic cables between 16 to 17 percent, said Frank Eickhorn, product manager for fire detection at Wagner Alarm and Security Systems GmbH in Hanover, Germany. Wagner makes electric compressors that use a special membrane to remove some of the oxygen from the outside air, a system the company calls OxyReduct. The excess oxygen is exhausted, and the remaining nitrogen-rich air is pumped inside the data center."
This discussion has been archived. No new comments can be posted.

Data Centers Breathe Easier With Less Oxygen

Comments Filter:
  • by drinkypoo ( 153816 ) <> on Monday March 19, 2007 @05:18PM (#18406269) Homepage Journal, so to speak. But it can't hold a candle to the burning excitement of watching pasty-faced geeks burn out, run out of steam, and pass out in a low-oxygen environment.
    • by Anonymous Coward on Monday March 19, 2007 @05:23PM (#18406355)
      Just imagine the new employee first day:
      - Here is your cube
      - Here is your chair
      - Here is your scuba gear ...
      • Re: (Score:2, Informative)

        by Anonymous Coward
        Make that an SCBA (like a firefighter wears), the 'U' in Scuba stands for underwater. ;)
      • by Dogtanian ( 588974 ) on Monday March 19, 2007 @05:43PM (#18406623) Homepage

        Just imagine the new employee first day
        I can see this whole process being abused by somewhat amoral bosses:-

        Boss (on telephone to sysadmin in data centre): "I'm sorry Dave, but your recent conduct just hasn't been acceptable. I've decided to invoke the disciplinary procedure, and having discussed this with Mr. Flibble we've decided that this warrants 2 hours of W.O.O."
        Sysadmin: "What's W.O.O.?"
        Boss: "With ... out ... oxygen. No oxygen for 2 hours. That'll teach you to be a git."
      • by Trillan ( 597339 )
        No, the first day would be devoted into a hazing ritual: mainly, making him pass out in the low oxygen room just for laughs.
    • by sczimme ( 603413 ) on Monday March 19, 2007 @05:38PM (#18406565)

      But it can't hold a candle to the burning excitement of watching pasty-faced geeks burn out, run out of steam, and pass out in a low-oxygen environment.

      Watch an out-o'-shape pasty-tubby try to ride a bicycle some time: with all his belabored breathing, one would think he was climbing Everest instead of pedaling on level ground.

      I, of course, am in perfect shape, with nary an ounce of extraneous tissue to be seen...

      *looks around furtively*
      *runs away*
      *collapses after 30 yards*

      • Well, I admit I resemble the fatass remark myself, although not to TOO horrible a degree since I'm fuckentall. But I also have asthma which makes it difficult to exercise. I can usually hike, so long as I don't want to do any talking at the same time (I'm too busy wheezing.) I mostly just need to get a good set of free weights, because that's easier to start and stop and isn't quite so aerobic. And go back on the Atkins diet, which I'm doing after my birthday at the end of the month.
        • by sczimme ( 603413 )

          No worries! I probably would collapse after 30 yards. :-) I ruptured my Achilles tendon *mumble* years ago and have been on a rigorous schedule of sitting ever since. Come springtime I'll be the wobbling wide-load on the bicycle!

    • That happened in Alcan Fabrications (banbury) in the late 1970's.

      They had a fancy nitrogen based system to put out fires, and it was tripped by accident, almost doing for a few of their technical bods.

  • by spun ( 1352 ) <> on Monday March 19, 2007 @05:18PM (#18406281) Journal
    Hehe, I can just picture Simon locking someone in one of these and slowly dialing down the oxygen until he gets that raise or perk or whatever he's after.
    • "Hostage" (Score:3, Funny)

      by HTH NE1 ( 675604 )

      "Shut up. We must conserve the air for as long as possible."
      "How long have we got?"
      "How many?"
      "I'll let you know."
  • Isn't this how halon systems work? It binds with the oxygen to make some other chemical and thus reduces the amount available for combustion?
    • Re: (Score:2, Interesting)

      by Anonymous Coward
      It doesn't bind to the oxygen so much as just displace it. Halon is heavier than oxygen and just pushes it out of the area. Halon is dangerous though because if there are people in the room when it goes off they won't be able to breath.

      This whole idea doesn't seem that great. So what if something shorts out and sits there glowing red and no one notices? You sure as hell notice when something starts burning but something could be slowing frying multiple components before anyone notices because there woul
      • by nietsch ( 112711 ) on Monday March 19, 2007 @05:34PM (#18406515) Homepage Journal
        The lower oxigen content just means that fires will not selfsustain. But if you have an external source for energy input, like the short you mentioned, thngs will still get hot and start to smoke. The chances are just a bit better that it does not cause a full-on fire.
        You are not supposed to be working all the time in the serverroom anyway, it's much too noisy in there and your 200Watt of heat production would be much better used to warm your office.
        In other words: you would have noticed that fire too late anyway if you had to rely on the amount of smoke coming from it.
        • Re: (Score:3, Interesting)

          by falcon5768 ( 629591 )
          HAHAHAHAHAHAH I wish I could quote this and give it to my boss....

          Not only do I work in the server room... we have CUBICLES in the server room with the network admins office RIGHT NEXT TO THE STACK. Needless to say, I cant hear the secretaries phone ring anymore.

      • It's not that they won't be able to breath, they'll breath just fine, except that the oxygen in their air has been displaced by HALON, meaning their air has not enough oxygen to sustain human life. Which is all well and fine because USAF flight crew tell us that losing consciousness due to lack of oxygen is very comfortable. If the HALON gas flows across something very hot (glowing metal perhaps). Then the HALON decays into some nasty chemicals akin to mustard gas. Oop's best to get out and stay out. T
    • Yeah but Humans need oxygen too you know.

      Halon systems eliminate basically all oxygen.
      These systems just reduce the amount of oxygen for the same effect.
      • >>> Yeah but Humans need oxygen too you know.

        Of course you're working off the assumption the people in IT are human....
    • Re:Mechanical Halon? (Score:5, Informative)

      by Big Bob the Finder ( 714285 ) on Monday March 19, 2007 @06:06PM (#18406925) Homepage Journal

      Halons work to extinguish fire using several mechanisms. Oxygen displacement- not absorption or binding- is one of them, but if this were the only factor, then dry nitrogen, carbon dioxide, or other inert gas would work just as well.

      There are four things required for combustion: oxidizer, fuel, heat, and a chemical reaction that is self-sustaining- the "chain reaction," in which free radicals are formed. Halons work by kicking off chlorine, bromine, or fluorine radicals in the heat of the fire, ending these reactions. Unfortunately, the same properties that make this class of compounds so wonderful for extinguishing fires is also what makes them so good at terminating the production of ozone.

      I also seem to recall something in my distant past as a fire instructor that halons as a group have a fairly high specific heat, meaning they carry away more heat from the fire; this is a relatively minor factor when compared to things like water which have high specific heat and very high heat of vaporization. Water is surprisingly good at putting out electrical fires; energized systems can be handled by using distilled water, as was done at Browns Ferry nuclear power plant in Tennessee in 1975. But it's messy and doesn't fight "three dimensional" fires very well.

      Replacements such as FM-200 and Novec 1230 that do not survive long enough to reach the stratosphere have been made and are now available. They are comparable in effectiveness to more traditional halons (Halon 1211 and 1301), and Novec is shipped as a liquid rather than a compressed gas. This makes it safer and less expensive to transport. Being fluorinated molecules (no chlorine, just fluorine) less phosgene is produced during a fire, which is a good thing.

  • Safe to work (Score:2, Interesting)

    by stanmann ( 602645 )
    Yes, its safe to enter, but how long, 1 hour, 3 hours 6 hours 8 hours. The article doesn't mention.
    • Re:Safe to work (Score:5, Informative)

      by scheme ( 19778 ) on Monday March 19, 2007 @05:28PM (#18406407)

      Yes, its safe to enter, but how long, 1 hour, 3 hours 6 hours 8 hours. The article doesn't mention.

      RTFA, the oxygen content in the air would be the same as living at around 2000-3000m which people certainly do without ill effects.

      • by ad0gg ( 594412 ) on Monday March 19, 2007 @05:52PM (#18406745)
        Not only can it prevent fires but it also help systems administrators train for the olympics.
      • by pla ( 258480 )
        RTFA, the oxygen content in the air would be the same as living at around 2000-3000m which people certainly do without ill effects.

        Of course, those people who live at 2000-3000m use fire for staying warm at night and for cooking. Kinda makes the entire premise seem a tad like... Oh, I dunno, complete and utter BS?

        That said, I've often wondered why we don't do/keep almost all dangerously flammable tasks/equipment in near-zero-oxygen environments. Seems like a bit of a no-brainer for situations requir
        • Re:Safe to work (Score:4, Interesting)

          by SEE ( 7681 ) on Monday March 19, 2007 @07:28PM (#18407965) Homepage
          Your lungs are mostly worried about the partial pressure of oxygen; .16 bars is what you need. Your lungs don't care too much if that's .16 bars of 100% oxygen, or one bar of 16% oxygen, or two bars of 8% oxygen. The level of concentration of oxygen doesn't matter too much, just the pressure of oxygen to drive membrane gas exchange.

          Fires, however, do not have gas-exchange membranes like your lungs, making the partial pressure less important, and the concentration more so. 8% oxygen at two bars is less supportive of fires than 100% oxygen at .16 bars.
    • Re: (Score:2, Interesting)

      by Anonymous Coward
      Well, it does say that it's like being at an elevation of 6000 ft, which is a perfectly habitable environment indefinitely. Takes some adjustment, but basically harmless.

      But, that must be making some assumption about the actual elevation of the datacenter. If the datacenter really is at 6000 ft. (it would be close to that, for example, in Denver, CO), then what is the effect of the reduced O2 concentration? At what point do you have to pressurize your datacenter to make the reduced O2 concentration safe?
    • Just your memory doesn't function as well, so you better make all the passwords really simple.
  • TFA is way too complex. There are much simpler ways to handle the problem. The oxygen levels in many major cities are below 18% already. Just let CO2 levels keep going up, this will push oxygen percentages down a tad more, and we have no more computer fires.
    • Troll? It's a joke for chrissakes! Sheesh!

      Not that it's funny, but come on moderators, lame or not, it's still a joke! Lighten up!
  • This should also help keep cleaning personnel out of the inner sanctums of the datacenter [], and therefore prevent downtimes due to accidentally plugged-out cables and stuff. And even in case it fails to keep them _out_, it might keep them _inside_ for a loooong time. Relativley well-preserved.

    I'm such a morbid bastard at times :/
  • Oh great! (Score:4, Funny)

    by zappepcs ( 820751 ) on Monday March 19, 2007 @05:23PM (#18406351) Journal
    Now mountain climbing, hang gliding, and other low oxygen sports will be important on my resume!!
  • I just know that The BOFH [] is going to be getting one of these systems installed soon. Only his system will occasionally reduce the oxygen levels much further than is strictly necessary for fire protection.

  • Sorry, not really datacenter related but it's brought up in TFA. So, at 6000 ft oxygen is much lower similar to this new system. So, there are no fires at that height? Is this true? How about off-site datacenters in the mountains (by the ski slopes)?
    • by JDevers ( 83155 )
      And ample cooling to boot.
    • by Drawkcab ( 550036 ) on Monday March 19, 2007 @05:35PM (#18406533)
      Its not exactly the same as being at 6000ft, its just similar from the perspective of how easily a human can breath. Higher altitudes have the same percentage of oxygen in the air, they just have lower air pressure, meaning less of all of its components. The lower altitude air will still be higher pressure, but with less oxygen. In terms of breathing, we just care about the partial pressure of oxygen, but thats not all that matters when it comes to whether something will burn.
  • by solevita ( 967690 ) on Monday March 19, 2007 @05:29PM (#18406429)
    Fire needs oxygen. More on this one as it comes in.

  • Not only does it stop fires, but it gets rid of your stupid employees!
  • by G4from128k ( 686170 ) on Monday March 19, 2007 @05:33PM (#18406495)
    Although I'm sure this is safe for day-to-day operations (for low-altitude data centers) and will prevent a self-sustaining blaze, I'd bet that a smoldering powersupply would convert an unpleasant fraction of the low-oxygen atmosphere into carbon monoxide. Oxygen-staved combustion tends to produce this deadly gas (which kills by binding to hemoglobin better than does oxygen)
  • If you didn't care about cost and or keeping people alive in the data center Helium would be the ideal inert gas.
    No fires to worry about and it is a great conductor of heat.
  • Poisonous exhaust (Score:3, Insightful)

    by youthoftoday ( 975074 ) on Monday March 19, 2007 @05:38PM (#18406569) Homepage Journal
    We had a similar issue when with the proliferation of large power-stations: water was pumped into cooling towers and then dumped in rivers. The cooling process de-oxygenated the water and this obviously meant the 'poisoning' of rivers (fish unable to breathe etc). We have a similar situation here. Only this time, the facility actually holds on to the oxygen. Why not mix it with the exhaust air (I'm sure it's not completely recirculated?) and avoid the potential for a similar situation. I know TFA says it's beathable, but it's worth considering the option nonetheless. Not all animals are humans. Remember what scale datacentres operate on, and which direection they're going in (they're not getting smaller). Has the potential not to be a significant issue...
  • What fun (Score:5, Funny)

    by Experiment 626 ( 698257 ) on Monday March 19, 2007 @05:39PM (#18406577)

    Not only are server rooms windowless, freakishly cold, and with uncomfortable chairs, but now they asphyxiate you too.

  • by tsstahl ( 812393 ) on Monday March 19, 2007 @05:44PM (#18406659)
    Redesignate the open floor space as the management conference room. The oxygen will be sucked out in no time.
  • Great (Score:4, Funny)

    by GreggBz ( 777373 ) on Monday March 19, 2007 @05:52PM (#18406751) Homepage
    Great, now I have to wear a wireless bluetooth headset AND an oxygen mask when I'm on a tech support in the Data Center.
    The guys in HR already call me "space man."
  • many many moons ago, i worked briefly at an oxygen concentrator manufacturing company. which is basically what this unit sounds like, an oxygen concentrator that sort of works in reverse, you keep the exhaust and throw away the product. these devices work by forcing compressed air through a molecular sieve - nitrogen adheres to the sieve and O2 passes straight through. but then you've got to get the nitrogen back out of the sieve, which requires decompressing the filter medium. all that energy you used
  • by Chyeburashka ( 122715 ) on Monday March 19, 2007 @06:06PM (#18406921) Homepage
    A few decades ago I served on a submarine. The oxygen generator stopped working for a while, and for operational reasons we couldn't snorkel for fresh air. The percentage of oxygen dropped below the point where combustion is supported, so the smokers were out of luck. People's lungs respond to the partial pressure of oxygen in air, not the absolute percentage, so the crew including myself were fine, since we were only at about the equivalent of 10,000 feet (US units). I always wondered wouldn't it be safer from a fire prevention standpoint to always operate like that.
    • by jonwil ( 467024 ) on Monday March 19, 2007 @09:37PM (#18409175)
      More to the point, why are they even allowing smoking in the closed environment of a submarine in the first place?
      • Re: (Score:3, Interesting)

        by Just Some Guy ( 3352 )

        More to the point, why are they even allowing smoking in the closed environment of a submarine in the first place?

        No kidding! Can you imagine the kind of power it'd take to drive the air scrubbers for something like that? You'd have to plug them directly into a nuclear powerplant or something.

        The reasons they'd allow smoking in that environment are that 1) in the scheme of things, the extra cleaning capacity needed to get the smoke out of the air is trivial, and 2) submarines are not exactly conducive

  • Imagine your glowing red hot but not quite burning cable inside a low oxygen cabinet. The equipment isn't working well, some some poor tech is sent to fix it. Said tech opens the cabinet, introducing a lovely fresh mix of 21% oxygen into the cabinet, at which point the superheated pyrolized gasses mix with the oxidizer and you get what we in the fire department like to's very bad for the complexion.

We declare the names of all variables and functions. Yet the Tao has no type specifier.