[ZODB-Dev] how big can a ZODB get?

Andreas Jung lists at andreas-jung.com
Wed Jan 12 14:02:20 EST 2005



--On Mittwoch, 12. Januar 2005 13:44 Uhr -0500 Tim Peters <tim at zope.com> 
wrote:

> [Victor Ng]
>> Thanks for the info Jim, where can I find information on ZODB tuning? I'm
>> running ZODB 3.3 right now and I'm not sure where to start looking for
>> the 'knobs' I need to turn.
>
> As Jeremy implied, app and object design is much more important than...

Object design is very important as Tim says. As an example I have seen code
(and written code :-)) where a large amount of data has been stored as 
attribute
of an object, e.g.

   self.filecontent=open('my_very_large_file').read()

which looks pretty straight-forward but shows a bad behaviour with ZEO.
Instead an implementation using a linked list of data chunks as in 
OFS.File.py
should be used instead.

-aj



More information about the ZODB-Dev mailing list