Re: [Hampshire] [OT] MTBF

Top Page

Reply to this message
Author: James Courtier-Dutton
Date:  
To: Hampshire LUG Discussion List
Subject: Re: [Hampshire] [OT] MTBF
2009/7/18 Andy Smith <andy@???>:
> Hi James,
>
> On Sat, Jul 18, 2009 at 10:44:49PM +0100, James Courtier-Dutton wrote:
>> Take an example of a SSD
>> Manufacturer quotes:
>> OCZ-120GB-Solid-State-Drive: MTBF 1.5 million hours.
>> To me that means and MTBF of 171 years abouts.
>> So, why do OCZ only guarantee it for 2 years?
>
> MTBF figures for hard drives (and possibly other things) are quoted
> in terms of analysis of large numbers of that model.  The item is
> rated for 2 years of use.
>
> What they are saying is that if you got many of these items and ran
> them simultaneously for 2 years then you should get 171 years
> aggregate service out of them before any of them developed a fault.
>
> Or if you ran one item for 2 years then replaced it and kept going,
> it should be 171 years before you had a failure inside the 2 years
> of any individual item.
>
> Cheers,
> Andy
>


To get a MTBF figure:
Approx MTBF = ( Number of Devices tested * Test time ) / Number of failures.
So, you are right, there is no indication as to what might happen after 2 years.
The model for failures is a U curve anyway. In that some failures
happen early on due to burn in etc., then it is fairly quite for a
bit, and then the failures start happening again at an exponential
rate.
This link helped me
http://www.vicr.com/documents/quality/Rel_MTBF.pdf