thanks for the info, and i'm already well aware of the commit options and the dbindex 
xml-element. those certainly can improve performance.

but i believe those things just pertain to the database or primary-key caching itself, 
right, and not to the caching behaviour for finder methods in particular? for 
instance, as the other gentleman pointed out, he found the finder method was always 
loading in all his 1/2 million primary keys! what i'm specifically wondering is if 
caching for finder methods can be made just as efficient (i.e., all in-memory) as it 
is for caching by primary key. 

basically what i'm imagining (in lieu of investigating the actual code) is that the 
jboss cache is fundamentally a big [in-memory] Map from recent primary keys to entity 
bean objects, obviating the need for serialization and database access for those keys 
in the Map. 

so, "is there also a Map of recent finder-parameters to collections of entity bean 
objects?" 

that would be the 'finder cache' i'm looking and hoping for. 

thanks again, mike


View the original post : 
http://www.jboss.org/index.html?module=bb&op=viewtopic&p=3823460#3823460

Reply to the post : 
http://www.jboss.org/index.html?module=bb&op=posting&mode=reply&p=3823460


-------------------------------------------------------
SF.Net is sponsored by: Speed Start Your Linux Apps Now.
Build and deploy apps & Web services for Linux with
a free DVD software kit from IBM. Click Now!
http://ads.osdn.com/?ad_id=1356&alloc_id=3438&op=click
_______________________________________________
JBoss-user mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/jboss-user

Reply via email to