Comments on: Intel study shows no effect from using none conditioned air http://www.openxtra.co.uk/blog/intel-study-shows-no-effect-from-using-none-conditioned-air/ Your daily IT cuppa Mon, 08 Mar 2010 00:45:34 +0000 http://wordpress.org/?v=2.9.2 hourly 1 By: Jack Hughes http://www.openxtra.co.uk/blog/intel-study-shows-no-effect-from-using-none-conditioned-air/comment-page-1/#comment-6865 Jack Hughes Mon, 29 Sep 2008 14:40:59 +0000 http://www.openxtra.co.uk/blog/?p=763#comment-6865 During the study the servers were placed in the car park (under cover one would assume) so I expect fire regulations weren't an issue. Data centres are sealed because of fire regulations as much as anything else. It is hard to see that changing any time soon. It would save a lot of power if instead of running the data centre at an ambient 18-21 celcius the temperature was permitted to be higher. In the study at least the operating temperature varied quite a lot apparently without any significant reduction in reliability albeit over a relatively short study. It seems that you can solve the power consumption problem in one of two ways: 1) cool the servers using much less electricity; 2) remove the requirement to cool equipment so much in the first place. Option 1 doesn't seem very likely any time soon unless those much heralded fusion reactors start working soon. :) During the study the servers were placed in the car park (under cover one would assume) so I expect fire regulations weren’t an issue. Data centres are sealed because of fire regulations as much as anything else. It is hard to see that changing any time soon.

It would save a lot of power if instead of running the data centre at an ambient 18-21 celcius the temperature was permitted to be higher. In the study at least the operating temperature varied quite a lot apparently without any significant reduction in reliability albeit over a relatively short study.

It seems that you can solve the power consumption problem in one of two ways: 1) cool the servers using much less electricity; 2) remove the requirement to cool equipment so much in the first place.

Option 1 doesn’t seem very likely any time soon unless those much heralded fusion reactors start working soon. :)

]]> By: Matt Simmons http://www.openxtra.co.uk/blog/intel-study-shows-no-effect-from-using-none-conditioned-air/comment-page-1/#comment-6864 Matt Simmons Mon, 29 Sep 2008 14:13:48 +0000 http://www.openxtra.co.uk/blog/?p=763#comment-6864 I don't see this as being much different from those occasional reports that come out from physiologists who claim that athletes don't need to stretch before a game. Great, interesting, but every coach in existence is still going to do it because they've had personal experience that not stretching leads to pulled muscles, just like most IT people have experience that says no AC is a <a href="http://standalone-sysadmin.blogspot.com/2008/06/infrastructure-upgrades-through-forest.html">Bad Thing(tm)</a>. I don’t see this as being much different from those occasional reports that come out from physiologists who claim that athletes don’t need to stretch before a game.

Great, interesting, but every coach in existence is still going to do it because they’ve had personal experience that not stretching leads to pulled muscles, just like most IT people have experience that says no AC is a Bad Thing(tm).

]]>