« IPv6 is Around the CornerShared Colocation Pricing Change »

1 comment

Comment from: Matthew Cooke [Visitor] · http://unrulymedia.com
Interesting points.

I do remember a study done a while back that showed that hard disks operate most reliably at around 40 degrees celcius and have high failure rates at low temperatures.

It's been a while since I hosted servers myself but I know when we went significantly above room temperature machines started failing very fast. I had always assumed if you were significantly below room temperature you'd be better off but I won't be assuming this in the future!

Incidentilly, I'm not sure if companies like Google are concerned with failure rates in quite the same way as small/medium sized companies, in particular they often don't prioritise single server reliability in the same way as smaller companies.
As an example I believe Google sometimes run CPU's overclocked and very hot. They don't care if this burns out the CPUs more quickly because they are better off getting as much bang per buck as quickly as possible given that in a years time the original CPUs will be looking pretty slow anyway.
02/23/09 @ 08:05

Leave a comment


Your email address will not be revealed on this site.

Your URL will be displayed.
(Line breaks become <br />)
(Name, email & website)
(Allow users to contact you through a message form (your email will not be revealed.)