Re: WAPBL vs. lfs?

["Michal Suchanek" <hramrach%centrum.cz@localhost>]

On 25/08/2008, Thor Lancelot Simon <tls%rek.tjls.com@localhost> wrote:
> On Mon, Aug 25, 2008 at 05:32:26PM +0200, Joerg Sonnenberger wrote:
>  > On Mon, Aug 25, 2008 at 11:20:33AM -0400, Thor Lancelot Simon wrote:
>  > > Regardless of the details (about which I believe you're mistaken) it is
>  > > still the case that, writing a continuous stream of directory or small
>  > > file creations, WAPBL does less than half the maximum throughput of the
>  > > underlying drive, while LFS can get close to 100% and is generally well
>  > > above 75%.  In normal operation there are no "dependency cycles" to break
>  > > in LFS.
>  >
>  > I'm not arguing relative to LFS. The comment was entirely about the
>  > second part, comparing WAPBL with plain UFS or softdep.
>
>
> The real fix for the "writes metadata twice" issue is to support a journal
>  on another partition.  With a real SSD (that is, one that's not built from
>  flash memory with its insane write penalty) this would make the extra
>  writes almost free.
>

It would also remove the stress form the disk area containing the
journal. My hardrive (although with ext3 filesystem) has just failed
due to bad blocks, and the fact some of them were in the journal are
is likely no coincidence. The hardrive surface wears off slower than
some early flash chips but it still wears off, and the areas that fail
first are things that are written the most - like the journal.

Thanks

Michal