Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations waross on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Server MTBF

Status
Not open for further replies.

i98davsc

Industrial
Apr 21, 2003
1
US
I'm building a TCO model for servers and I'm having some issues figuring out the maintenance effort. If I could determine the MTBF for a server, that would pretty much solve my problem. I talked to a guy at IBM who told me that they don't determine any such number for their machines - the MTBF figures usually stop at the component level. And honestly, what I'm looking for is the practical failure rate, not the 750'000 hours that you typically find in the manufacturer's specs.

So what I'm looking for is your practical experience: What would you say is the typical failure rate/year for low-end, mid-range, and high-end servers?
 
Replies continue below

Recommended for you

Just a caveat, it depends very heavily on how "abused" the server is.

MTBF is exponentially proportional to 1/T, where T is absolute temperature in kelvins. I've seen servers locked up in a broom closet and others in a nice air-conditioned, raised-floor room.



TTFN
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor

Back
Top