The 100 Degree Data Center
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."
Its better
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.
I realize it's the trendy thing these days to target the data center as an area of concern monetarily, but this is a little ridiculous.
All it will take is one poor geek spending a 12 hour day in the data center for this to be deemed a horrible idea. (Like that never happens)
Seriously, this is retarded. If you do your cooling and power CORRECTLY, you won't have a ridiculous bill and your data center will be at a more reasonable temperature.
I hate really hot weather...you can always put on more clothes, but you reach a limit on what you can take off.
Sent from your iPad.
"...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."
Not welcome? That all depends, on if I can relocate my Data Center to a topless beach in Miami. Sure beats the current scenery, and the dress code would likely change.
Of course, the fact that you probably don't want to see your average IT person running around topless wearing a thong is another matter entirely...
Buildings provide hot water for washing hands etc. Cold water comes in from outside and is heated using electricity or gas to make hot water which costs money and energy.
Pipe the cold water (which is usually somewhere between 0 and 20 degrees C) through heat exchangers in the hot data centre before heating it up to working temperature with gas or electricity.
That way, you reduce the data centre's temperature to more like 20-25C, and you heat the water up by 10C (say) saving on gas or electricity bills since there is less of a temperature difference to get it up to the required temperature.
I eagerly await my Nobel Prize for Common Sense.
Stick Men
The proper question is "Are our coworkers ready to deal with how we'll smell like after spending time in that server room?" It'll smell like a monkey house, but probably with less feces. Unless we're working with that superstar bastard programmer a few articles back who poo'd in the lobby.
Kwisatz Haderach
Sell the spice to CHOAM
This Mahdi took Shaddam's Throne
We upped the temperature in our small data center to 75~80. Those systems in there run just fine at around (and a little higher then) room temperature. I didn't really see any need to keep it running like a refrigerator for no reason. The AC runs less, there must have been some money saved, but it is more comfortable in there the few times I have to do something there.
"I use a Mac because I'm just better than you are."
welcome our new sweaty sysadmins.
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.
How about completely immersing the entire data center in oil? You may need some sort of SCUBA gear to to server maintenance though.
Data center is hot, racks are cool. Hot contents of racks make data center hot. Computer not like 100 degree heat, computer fail
It's not like water cooling is new.
Yes there saving money on cooling cost, or at least they seem to believe that and I am sure when they fail to take everything into account this is true.
The reality it the server room still has to pull that heat out. Increased Delta T is just lost energy.
Here is really why it's a terrible idea.
1.) Component failures. Of all parts from bearing in the drives, and fans to the silicon itself has a much higher failure rate.
2.) The components use more power at higher temperatures! This is from increased leakage currents in the silicon.
Below is a graph from Research My Startup company did!
http://www.silentcomputing.com/tech/market2.gif
They really need to used ducted air or any other technology to reduce the Delta T! By this I mean bring the cooling as close to the components as possible.
Right now server rooms need to run internally at 10C to 15C to keep the CPU chips below 60C.
If they just brought the cooling directly to the cpu's and let that cool spread from there they could use out door passive radiators! 0 air conditioning cost and the most power savings.
This is what my start was doing till someone tried to steal the who damb thing and sunk the company.
I am always doing that which I can not do, in order that I may learn how to do it. - Pablo Picasso
That might work for 6 months of the year, but we get these things called summers... We can hit 100 degrees or more here in Washington State. Even Alaska gets warm in the summer (Fairbanks record high of 99F).
Real programmers use "copy con program.exe"
373.15 - Water boils at 1 atmosphere 310 - Very hot 300 - Hot 290 - Nice 280 - Chilly 273.15 - Water freezes at 1 atmosphere 0 - absolute zero! how easy is that.
'xactly. These trolls try to go anti-US with their fancy metric system then they fuck it up with Centigrade. Try plugging centigrade temperatures into the ideal gas law and lemme know how it goes. ;)
A single degree Celsius is qualitatively a bit too big, to the point where most European climate-control systems with digital displays have to resort to using half-degrees as the base control unit.
10 PRINT CHR$(205.5+RND(1)); : GOTO 10
I for one welcome our new sweaty sysadmins.
What are you going to do with the old ones?
I am literally 3000 tokens away from the chaotic crossbow --Stephen
The metric system is unified in all directions, time, mass, length, temperature, energy etc...
The system makes sense instead of relying on the length on the king's thumb, foot and arm, or the weight of a stone or the amount of work being done by a horse, all variable and inconsistent.
Its one of the many things we owe the French under Napoleon, like a unified system of laws, the "Code Napoleon."
MSBPodcast.com The opinions expressed here are my own. If you don't like 'em... Think up your own stuff.
SUN already beat you to it... /. when it first came out...
http://www.sun.com/products/sunmd/s20/specifications.jsp#Anchor6
It was also posted on
...but the servers aren't the only thing in a data center. If the switches and routers can't take the higher heat, then you aren't going to get much use out of those servers.
If you can't beat them, embrace and extend them.
The real benefit to a 100F setpoint is the free cooling it allows. You can use filtered out door air or evaporatively cooled water from a cooling tower to keep a datacenter at 100F year round just about anywhere. This is a 90% reduction in cooling energy right there using decades old, mature HVAC tech.
May I point out the obvious: not only the higher power consumption comes from increased leakage currents in the silicon, but it also comes from the fact that power supplies are less efficient at higher temperatures, so they need to pull more current from the wall socket to maintain the same output current.
However what you and I just said is irrelevant. As your graph shows, the difference in power consumption is very minimal: 2% for each 10C due to leakage currents, and maybe ~5% for each 10C in decreased PSU efficiency. These few percentage points are nothing compared to the amount of power you would save by making the AC work less hard. Indeed, if without AC the datacenter would reach 140F (333 Kelvin), cooling it down to 60F (289 Kelvin) requires removing 44 Kelvin of heat, whereas cooling it down to 100F (311 Kelvin) only requires removing 22 Kelvin of heat, therefore running it at 100F would roughly reduce the AC power consumption by 50% ! So the point made by TFA still holds: overall you still are saving energy by running a whole datacenter at a 10C higher temperature.
As to the higher component failure rate: as it was proven by 2 independent studies last year (Google and CMU), higher temperatures do not even correlate with higher hdd failure rates. In fact, strangely they observed a slight reverse effect: hdd tended to fail less often !