The 100 Degree Data Center 472
miller60 writes "Are you ready for the 100-degree data center? Rackable Systems has introduced a new enclosure that it says can run high-density racks safely in environments as hot as 104 degrees (40 degrees C), offering customers the option of saving energy in their data center. Most data centers operate in a range between 68 and 74 degrees. Raising the thermostat can lower the power bill, allowing data centers to use less power for cooling. But higher temperatures can be less forgiving in the event of a cooling failure, and not likely to be welcomed by employees working in the data center."
Drives (Score:5, Interesting)
I'd be mostly concerned about the lifespan of hard drives at these temperatures. The electronics can be easily made to tolerate heat, but drives are a weak link. The bearings and lubricants are especially vulnerable.
Re:Brrr... (Score:3, Interesting)
Re:Drives (Score:5, Interesting)
80 degrees (Score:3, Interesting)
We have an 80 degree data center. It's not particularly pleasant to be in (as you get buffeted by hotter winds coming off of power supplies), but we haven't seen any more failures than normal.
Re:Were nerds here... use the f'ing metric system (Score:2, Interesting)
Fahrenheit is stupid.
Wow, how insightful and deep. You have to think real hard to come up with such compelling commentary?
Celsius on the other hand is much easier to remember:
0 - Water freezes
10 - Cool
20 - Nice
30 - Hot
40 - Scorching hot
50 - Burn sensation
100 - Water boils
Surely you mean water freezes and boils at one standard atmosphere, right? Which brings me back to my point about it being just as arbitrarily defined as Fahrenheit was. Fahrenheit also offers more precision without using decimals.
And slashdot.org is not an american-only site as it's domain name ends in .org and not in .us
Domain names don't mean jack. Slashdot is American owned with a largely American readership. Yet someone still managed to whine about the fucking summary using American measurements, in spite the fact that the metric measurements were also provided. Hmm.......
Re:Brrr... (Score:1, Interesting)
At 100 you get into different lower cost ideas for cooling (in some parts of the country). Such as a fan and a few filters to filter outside air.
Such as taller roofs and large slow spinning fans to exhaust air.
All you have to do is keep excess moisture out.
Now these computers have to coexist with others that do not tolerate that sort of temp range. So the data centers still remain on the coolish side.
SUN's blackbox does this... (Score:2, Interesting)
http://www.sun.com/products/sunmd/s20/specifications.jsp#Anchor6 [sun.com]
It was also posted on
Re:Were nerds here... use the f'ing metric system (Score:3, Interesting)
Farenheight has no basis in anything practical at *any* range. At least Celsius is based around water, which is useful for a number of reasons.
That's not quite true, 0*F to 100*F is pretty much the range humans can survive in without any kind of crazy technology. Makes sense to me.
Re:Were nerds here... use the f'ing metric system (Score:2, Interesting)
Well to be fair, 0 F was supposed to be the freezing temperature of water when fully saturated with salt (I believe), though it turned out to be less than 0 F later. 100 F was supposed to be body temperature, which of course isn't quite right.
I like Farenheit because of the finer divisions per integer and the 0-100 intuition, but moreso because I grew up with it. Oh well.
Lets switch to metric time while we're at it, what's this 24/60 crap?
Re:Were nerds here... use the f'ing metric system (Score:2, Interesting)
I completely fail to see how a range of 40-80 (after all, you did say "habitable temperatures" for humans), is better than a range of 5-30.
Humans can survive in climates colder than 40F.
On the other hand, it is completely arbitrary to use Fahrenheit over Celsius. And vice versa. It does not matter one whit that ice melts at 0C or 32F in your daily life. If you actually needed a "logical" and consistent unit for temperature, you would be using the Kelvin scale.
What is so logical about the meter being defined as the length of a certain platinum bar? Nothing, it is just as arbitrary.
The SI system has one major advantage over the US Customary system. Unit conversions, in terms of scalar multiples, are far easier. And that advantage is not relevant to temperature units, in our daily lives. Something is seriously wrong in the world if you're worrying about mega-Celsius or mega-Fahrenheit temperatures outside of theoretical physics (where you would be using Kelvin anyway)
Re:Were nerds here... use the f'ing metric system (Score:5, Interesting)
Well, yeah, except for that 98.6 thing.
You're referring to the common misconception that that's the normal body temperature, or even that body temperatures are so regular that you'd need a decimal point to express it? (The figure 98.6F is an example of false precision, being a translation of 37C, which wasn't meant to be more accurate than a degree celsius to begin with, and was a rough figure at that.)
Re:Drives (Score:3, Interesting)
Here's the whitepaper [74.125.47.132] (HTML-converted).
I'm not able to open the PDF right now to see the pretty graphs, but it says "The figure shows that failures do not increase when the average temperature increases. In fact, there is a clear trend showing that lower temperatures are associated with higher failure rates. Only at very high temperatures is there a slight reversal of this trend." However, it also notes that "What stands out are the 3 and 4-year old drives, where the trend for higher failures with higher temperature is much more constant and also more pronounced."
They don't even mention free cooling (Score:2, Interesting)
Re:Centigrade sucks! (Score:3, Interesting)
Try plugging centigrade temperatures into the ideal gas law and lemme know how it goes. ;)
Actually, the fun part is that there's a long history of people doing exactly that. The patent offices in the US and other countries have an ongoing problem of people attempting to patent perpetual-motion machines. In most cases, a "proof" that a particular gadget will work is produced by taking the standard equations and using Celsius/Centigrade numbers when temperature in "degrees" is needed. This seems to be something that a lot of physics cranks can't quite get right. They read that the Kelvin degree is the same as the Celsius degree, and that means that it doesn't matter which you use, right?
Re:the larger degrees are nicer (Score:3, Interesting)
The US will never fully implement the metric system. And I mean NEVER (see my caveat below). It's an impediment to foreign companies supplying or working in our craft sector. Ever wonder why you don't see any European or Chinese sawn lumber or Chinese or European steel beams for purchase? Because they don't have factories that generate Imperial sizes. All the foreign factories are geared to the metric system. The only areas where I've seen a shift to metric was the car industry where foreign imports caused US manufacturers to start using metric parts and the plastic drink container industry where uniform packaging became a problem and they simply implemented metric sizes to negate the need for American specific sizing.
It's a perfectly legal impediment to open trade, much like many of the European ones that the US doesn't use. It keeps jobs and production in the US and (*)until the day the US stops importing more than we export it WON'T change. The government implemented a Metric requirement on Roadway design during the Carter administration. That requirement was completely waved at the end of the Clinton Administration because of the impact it would have had to the craft supply industries where they would have to compete against imported products subsidized by foreign governments.
The imperial measurement system keeps jobs in the US. It's serves a very important purpose beyond being what people know.