Most of these "optimizations" look very desparate to me; most likely
they are motivated by a very large number of rules. Perhaps some
techniques for reducing this number would show greater benefit.
-W
On 13/09/2012, JP Chemali <jshemali(a)hotmail.com> wrote:
Hi all,
I'm trying to reduce the memory footprint of packages in my application
(for
performance reasons they are dynamically loaded from disk), and I was
wondering if some of you had any experience with the subject?
So far the optimizations that I can come up with in 5.4.0.Final version are
all about reducing the sizes of generated classes:
- Package name in the drl file should be as small as possible, one letter
is
best (even better that no package at all)
- Rule name should be as small as possible as it is used to generate the
consequence class name
- Write as less code as possible in the consequence
- Use as few as possible eval statements
- By using my own JavaDialectConfiguration and MVELDialectConfiguration
when
building the knowledge, I can remove the random number on class names that
was added in the recent versions (don't understand why it was added in the
first place? Does anyone know?)
I see some leeway to shorten the names of the following generated classes
but the amount of overriding to change the "default" consequence name and
the suffixes is a bit too exotic for me:
- <PackageName>.<RuleName>DefaultConsequenceInvoker
- <PackageName>.<RuleName>DefaultConsequenceInvokerGenerated
Any advice, other leads some of you might have?
Thanks in advance
--
View this message in context:
http://drools.46999.n3.nabble.com/Package-footprint-tp4019721.html
Sent from the Drools: User forum mailing list archive at
Nabble.com.
_______________________________________________
rules-users mailing list
rules-users(a)lists.jboss.org
https://lists.jboss.org/mailman/listinfo/rules-users