Urg. simul.py was supposed to provide an alternative *schniff*
If theoretical hit rate is all that matters to you, yes.
Damn it! Where's my magic bullet? And where's that cake I was supposed to have AND eat? It was supposed to be here by yesterday...
More seriously, yes, I suspect you're entirely correct. In this case I suspect the enterprise-level-backed-up-disk-thingy interracting with directory storage's let's-create-millions-of-files-and-subdirs is probably having more of an effect, and even that is likely dwarfed by inefficiences in the app-level code...
If you're serious, you also need to play with changing the target number of
objects in your ZODB (Connection; in-memory; "pickle") caches.
It would be great if this could be set to a max-limit in MB. How hard do you reckon that'd be to implement?
-- Simplistix - Content Management, Zope & Python Consulting - http://www.simplistix.co.uk
_______________________________________________ For more information about ZODB, see the ZODB Wiki: http://www.zope.org/Wikis/ZODB/
ZODB-Dev mailing list - ZODB-Dev@zope.org http://mail.zope.org/mailman/listinfo/zodb-dev