> Now what would be cool is a journaled file system for bsd.  Any one out
> there?  Perhaps when bsd supports xfs, reiser or ext3.

I think soft dependences is basically BSD's answer to journaling. Here's an
exerpt from the NetBSD FAQs - http://netbsd.org/Documentation/misc/ :
---
Q Are there any measured improvements of softdep? (top)
  
  The  fact  that meta-data isn't immediately written to disk makes the system
  feel a lot faster. No numbers are available to show this at this moment.
     
  Paul  Vixie  has run a benchmark on storing and retrieving (in random order)
  75.000  files  into  a directory, on both a system with and without softdep.
  The  results show that file access time for both creating and locating files
  in large directories are generally better on systems with soft dependencies.
  (The softdep system in this case was FreeBSD 4.2, the non-softdep system was
  BSD/OS  3.1.  Similar  behaviour will be visible on a NetBSD system with and
  without softdep).
       
Q How resilient is the Fast File System (FFS), FFS with softdeps and NetBSD's ext2fs
  implementation with respect to unclean shutdowns? (top)

  The FFS takes care to correctly order all metadata operations, as well as to
  ensure  that all metadata operations precede operations on the data to which
  they  refer,  so  that  the  file system may be guaranteed to be recoverable
  after a crash. The last N seconds of file data may not be recoverable, where
  N is the syncer interval, but the file system metadata will be. N is usually
  30.
     
  With  softdeps running, you've got almost the same guarantee. With softdeps,
  you  have  the guarantee that you will get a consistent snapshot of the file
  system  as  it was at some particular point in time before the crash. So you
  don't  know,  as  you did without softdeps, that, for example, if you did an
  atomic  operation  such  as  a  rename  of  a  lock file, the lock file will
  actually  be  there;  but  you do know that the directory it was in won't be
  trashed  and  you  do  know  that  ordering dependencies between that atomic
  operation  and  future atomic operations will have been preserved, so if you
  are  depending  on  atomic  operations  to  control, say, some database-like
  process  (e.g.  writing  mail  spool files in batches, gathering data from a
  transaction  system,  etc.) you can safely start back up where you appear to
  have left off.
     
  NetBSD's ext2fs implementation gives you the traditional FFS guarantee about
  metadata  (unlike the Linux implementation), so you can actually use it with
  more confidence than you can use the native ext2fs in Linux. The downside is
  that  it's a bit slower, but that's because it actually does the right thing
  if  the  system  crashes,  instead  of  potentially  eating your file system
  itself.
---

Maybe someone more knowledgeable than I can comment on what "the right thing"
is refering to in that last paragraph.

In addition to using softdep one can mount portions of their system on
a memfs - basically a RAM disk. For a modern single-user system with lots
of memory this seems like it would be ideal, especially if using ECC memory.

Ref:  NetBSD Tuning: http://netbsd.org/Documentation/tune/5.html

-Beaker

P.S.  Thanks Neil - your last post regarding file descriptors answered
        my question as to how fd3-19 get "defined". 
--
[ SiMpLe MaChInEs ] --> gopher://beaker.mdns.org
   or (via proxy)
http://gopher.floodgap.com/gopher/gw.lite?gopher://beaker.mdns.org:70/1
_______________________________________________
EuG-LUG mailing list
[EMAIL PROTECTED]
http://mailman.efn.org/cgi-bin/listinfo/eug-lug

Reply via email to