Re: [tlug] SSD filesystem

[Francois Cartegnie <fcartegnie@example.com>]

Le dimanche 7 juin 2009, Clemens Schwaighofer a écrit :
> I guess they are all the same like HDs, who in the world wants to
> override the automatic badblock system in a normal HD?

Hi clemens,

No override for this, wear levelling and badblocks accounting/reindexing are 
totally different.

> FS need to change for that, so there will be some major rewrite to
> address the difference between oldschool HDs and SSDs, but that will
> take some time. SSDs are by fair a common good. Pricey for the good
> ones, and real crappy slow for the bad ones.

There's an opinion following linus recent remark about SSDs:
http://ldn.linuxfoundation.org/blog-entry/should-filesystems-be-optimized-ssd%E2%80%99s

I see more and more MLC devices appearing on the market with, or with more, 
RAM memory to improve performances (256..512MB...). That way you keep your 
frequent writes in ram (fs logs) and increase the reliability of those cheap 
dumb ssd devices for a low additional cost.

If you look back how RAID made it to the market instead of the specialized 
expensive replicated disks, you don't have the guarantee that priceys ones 
will be the answer.

And there's still a need for geeks (?) doing SSD disks with regular flash 
devices (compact flash, SDHC: you can even find multiple raid 0/1 controller 
for those ones). Most current embedded devices uses a read only FS for wear 
levelling reasons.

The kernel might need to end up with uniformization for SSD access, and SSD 
makers firmware to provide a way to tune parameters, the way you can do with 
scsi disks (or in a limited way with sata).

Francois