[svlug] More links about ssds
newmans at sonic.net
Thu Jun 2 13:16:15 PDT 2016
On 06/02/2016 11:38 AM, John Conover wrote:
> Since SSD memory elements are semiconductor based, it is probably a
> reasonable assumption that failures are ergodic, (including
> write/erase cycles.)
> MTBF does not mean how long a device would be expected to last.
> MTBF means that given sufficiently many devices, half would have
> failed by the MTBF, and half would be still be running, (this is NOT
> true if failures are non-ergodic.)
> If a system depends on multiple devices to function, (NOT a stripped
> array; SCSI, for example, which can tolerate at least a single
> failure,) doubling the number of devices reduces the system MTBF by a
> factor of two, (if failures are ergodic.)
I don't care about "RAID" 0, that isn't RAID.
The most recent paper I could find about SSD failure patterns is
https://users.ece.cmu.edu/~omutlu/pub/flash-memory-failures-in-the-field-at-facebook_sigmetrics15.pdf I haven't had a chance to digest it but looks
like they have found that the SSD failure pattern does not resemble the hard drive failure pattern at all.
More information about the svlug