I seem to remember a study from Google which indicated that drive failures = were loosely correlated to temperature and strongly correlated to manufactu= ring batch. --Michael > On Jun 11, 2014, at 9:22 AM, Harrison Cooper wrote= : >=20 > Interesting data, thanks for sharing. Some servers do a good job pushing= air, and it depends on where the drives are mounted of course. Ours usual= ly end up in the back side getting the full brunt of hot air from CPU's and= DRAM. Facebook actually puts ours in front, so we get the full effect of = the "cool" air. =20 >=20 > -----Original Message----- > From: piclist-bounces@mit.edu [mailto:piclist-bounces@mit.edu] On Behalf = Of Justin Richards > Sent: Wednesday, June 11, 2014 4:55 AM > To: Microcontroller discussion list - Public. > Subject: Re: [EE]:: Hard drive reliability >=20 > On 11 June 2014 05:24, Carlos Marcano wrote: >=20 >> I thought I should post this from Backblaze (please forbid if it has=20 >> already been posted): >>=20 >> < >> http://blog.backblaze.com/2014/05/12/hard-drive-temperature-does-it-ma >> tter/ >>>=20 >>=20 > Backblaze appear to be doing an excellent job keeping their drives cool. >=20 > It would be interesting to see failure rates over a wider range of temper= ature. That may prove to be an expensive exercise and suspect no one would= want to expose a significant number of their drives to higher temperatures= to make the data valid. >=20 > Would be interesting if they were willing to sacrifice some older pods an= d expose them to higher temperatures. >=20 > I am not sure I agree with "How much does operating temperature affect th= e failure rates of disk drives? Not much." >=20 > I think the only conclusion is:- not much over the temperatures theirs ar= e run at. >=20 > Justin > -- > http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive View/cha= nge your membership options at http://mailman.mit.edu/mailman/listinfo/picl= ist >=20 > This e-mail (and any attachments) is confidential and may be privileged. = Any unauthorized use, copying, disclosure or dissemination of this communi= cation is prohibited. If you are not the intended recipient, please notif= y the sender immediately and delete all copies of the message and its attac= hments. >=20 > --=20 > http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive > View/change your membership options at > http://mailman.mit.edu/mailman/listinfo/piclist >=20 --=20 http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist .