NetBSD-Users archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: state or future of LFS?



On Mon, Apr 13, 2009 at 2:24 PM, Andreas Gustafsson <gson%gson.org@localhost> 
wrote:
> Thor Lancelot Simon wrote:
>> Journalled FFS is not a replacement for LFS.
>
> Indeed it is not.  The main reason why I would love to see a working
> log-structured file system in NetBSD is that it seems like such a
> perfect match for solid state disks and other flash-based media.
> Flash tends to have performance problems with small scattered writes,
> which is exactly what LFS avoids, and if LFS has problems with
> fragmentation as some have claimed, it will benefit from a medium
> with very fast seeks, which is exactly what flash provides.
> --

This thread is an interesting deadlock between two parties:
Users who think LFS is a really good idea and want to see it thrive
versus Developers who have worked on LFS as it exists right now and
been frustrated/burned/otherwise made unwilling to continue working on
it

I fall into the first camp and think LFS would be great in FUSE:
http://mail-index.netbsd.org/tech-kern/2009/01/16/msg004052.html

However, the sheer number and caliber of developers who fall into the
second camp says a lot about the current implementation of LFS, so
maybe some middle ground would be a good place start.

I'll WAG a suggestion, but my own abilities really limit the amount of
weight I can put behind it:
Start an LFS-lite filesystem as a new project to help solve the
problems expressed by frustrated developers and targets the main ideas
keeping the LFS user community passionate about defending it.  (the
90% solution?)

Matt


Home | Main Index | Thread Index | Old Index