Re: nndb mailing list?

[Dave Blacka <dblacka@fuentez.com>]

>>>>> "Joe" == Joe Hildebrand <hildjj@fuentez.com> writes:

 Joe> Also BTW, we did some more testing on the differences between gdbm and
 Joe> berkeley db.  Gdbm wins *big* on disk space (like one less digit in
 Joe> the bytes column), and is not noticeably slower.  It will probably be
 Joe> the default in 0.8.  We'll also include a script to convert one to the
 Joe> other.

Ok.  Joe lied here a bit.  GDBM does actually win big on disk space:
Berkeley DB isn't really very space efficient when the data you are
storing has widely different sizes (like mail bodies), so the Berkely
DB file grows to enormous size.  For instance, I probably have at most
1 Meg of mail stored in a 7 MB mail.db file.  GDBM gets you much
closer to the actual size of the data.  However, GDBM and Berkeley DB
only have comparable speeds when you are hitting a local disk.  In my
testing so far, GDBM suffers from incredible performance problems over
NFS, while Berkeley DB behaves about the same.  Thus, if you are
storing your mail.db file on a local disk, switching over to GDBM is
probably a real good idea, otherwise it would probably be best to wait
until Joe and I rework the whole backend of nndb.

Assuming we can get all this to work, Joe and I will put all of the
message bodies in a separate database, and then write Tie:: modules to
implement different storage schemes.  The overview databases will
always be stored in a (separate) DB hash of some sort.

I'm guessing that the MH style storage scheme will probably be the
best and become the default.

Dave

-- 
David Blacka                    |dblacka@fuentez.com
Software Engineer               |Fuentez Systems Concepts