you're not going to believe this.
Wojciech Puchar
wojtek at wojtek.tensor.gdynia.pl
Wed Jun 24 09:44:42 UTC 2009
> First, be careful about statements like "10 years before it fails to hold
> state." Usually that means if you write data to the device and put it on a
> shelf, you've got 10 years before the data is unreadable. Being marketing
possibly it's true if you will write it few times and no more ;) store it
in perfect stable room temperature with low natural radiation background
etc.
> for a decade? The number you probably care about is how long _in active
> use_ the drive will last, and that's probably _not_ 10 years. The primary
10000 writes of it's size if it would be properly managed (flash
filesystem). As it emulates disk divide it by at least two on writing
large files, at least 20 in case of random small writes.
> is less noticeable. Implementing wear leveling in OS-level software
> isn't feasible. As I mentioned, wear leveling happens within the chip,
> so the OS doesn't even know a block swap has occurred. (As an extension
you are wrong. flash chips doesn't do this. There is a controller that do
this. If it would give simple interface to flash chips and say PCI-Express
or SATA, making proper flash filesystem would be possible.
More information about the freebsd-questions
mailing list