[Btrfs-devel] Re: btrfs and Solid State Disks (SSD).

[Chris Mason]

On Wednesday 13 February 2008, Andi Kleen wrote:
> Chris Mason <chris.mason at oracle.com> writes:
> > Yes, SSDs are a big target of mine, and so the parts that are not
> > currently favorable for SSDs will be changed.  The big problem right now
> > is that btrfs writes to a fixed super block for every commit.  That will
> > change to a rotating set of fixed super blocks to lower wear and improve
> > redundancy.
>
> The firmwares of the SSDs tend to do that anyways (except on embedded
> systems where the flash is directly connected without firmware, but that is
> probably not the right target for btrfs). So doing it in the FS is likely
> redundant.

We need super block copies regardless, and eliminating hot spots on the device 
is a good thing in general.  Even for traditional storage, frequently written 
sectors are the most likely to fail.

-chris