2:17 a.m.
Hello,
I am interested in running Drools 5 (and especially Drools Fusion) on a
>somehow<< resource-constrained device for which only a J2SE
1.4 VM ist
available.
Since Drools meanwhile requires J2SE 1.5 I thought about backporting it.
But before I invest so much effort, I got two questions:
1) I have downloaded the sourcecode. At a first glance the only reason for
requiering J2SE 1.5 is the extensive use of generics. In case I would
remove all generics from the code, would it be possible to compile drools
for a J2SE 1.4 VM runtime in general or are there other more hidden
reasons for requiering J2SE 1.5?
2) I already meantioned I am planning to run drools on a >>somehow<<
resourcen-constrained device. I explicitly used the term >>somehow<<
because I am talking of a device with a 400 Mhz CPU and 256 MB RAM (some
people might laugh about calling this resource-constrained). How about
resource-consumption of drools? Has anyone run drools on a device with
similar resources?
Thanks in advance!
Best Regards
Thomas Frenken
12:02 p.m.
New subject: Drools ReteOO versus Rete
Greetings:
Just wondering if this link still applies. If so, I think I may have
found an answer to one of my benchmark problems.
http://docs.codehaus.org/display/DROOLS/Rete+Vs+Reteoo
SDG
James Owen
Founder October Rules Fest
Senior Consultant / Architect KBSC
http://www.kbsc.com
http://www.OctoberRulesFest.org
Twitter: OctRulesFest
Blogs:
http://JavaRules.blogspot.com [Rulebased Systems Blog]
http://ORF2009.blogspot.com [October Rules Fest Blog]
http://exscg.blogspot.com/ [Expert Systems Consulting Group Blog]
"If I have seen a little further it is by standing on the shoulders of
giants."
Sir Isaac Newton in a letter to Robert Hooke, 5 Feb 1676
Come to October Rules Fest and stand on the shoulders of the Giants of
the industry; if only for a week.
On Aug 13, 2009, at 8:21 AM, mrhoden(a)franklinamerican.com wrote:
12:22 p.m.
New subject: Drools ReteOO versus Rete
James,
I believe this was related to Drools 2.
What exactly is your problem?
[]s
Edson
2009/8/14 James Owen <jco2009(a)att.net>
--
Edson Tirelli
JBoss Drools Core Development
JBoss by Red Hat @ www.jboss.com
12:48 p.m.
New subject: Drools ReteOO versus Rete
Edson:
When I run WaltzDB-16 I get only 9K+ rules fired rather than the
"normal" 21K+ rules fired. I'm thinking that IF this logic still
applies THEN that might explain why WaltzDB-16 does not fire the
requisite number of rules.
SDG
James Owen
Founder October Rules Fest
Senior Consultant / Architect KBSC
http://www.kbsc.com
http://www.OctoberRulesFest.org
Twitter: OctRulesFest
Blogs:
http://JavaRules.blogspot.com [Rulebased Systems Blog]
http://ORF2009.blogspot.com [October Rules Fest Blog]
http://exscg.blogspot.com/ [Expert Systems Consulting Group Blog]
"If I have seen a little further it is by standing on the shoulders of
giants."
Sir Isaac Newton in a letter to Robert Hooke, 5 Feb 1676
Come to October Rules Fest and stand on the shoulders of the Giants of
the industry; if only for a week.
On Aug 14, 2009, at 12:22 PM, Edson Tirelli wrote:
1:06 p.m.
New subject: Drools ReteOO versus Rete
No, I don't think that would be the problem. Considering that we always
calculate the full cross products/matchings, maybe that is due to a
different implementation of the conflict resolution strategy (assuming the
rules are all correct)?
Did you checked the number of canceled activations?
[]s
Edson
2009/8/14 James Owen <jco2009(a)att.net>
--
Edson Tirelli
JBoss Drools Core Development
JBoss by Red Hat @ www.jboss.com
1:35 p.m.
New subject: Drools ReteOO versus Rete
Edson Tirelli wrote:
Probably more importantly,
are you able to check that the actual correct
results where produced. I don't know of any published benchmarks where
the results have been properly verified, except checking the number of
rules fired.
Mark
8:26 p.m.
Ha - yeah how long ago would a machine like that have not been
"resource constrained" ! :)
But if it really is resource consttrained, some kind of static code
generation and pre compilation could help. Such tricks aren't uncommon
for phones, for instance.
Sent from my phone.
On 13/08/2009, at 11:21 PM, mrhoden(a)franklinamerican.com wrote:
5704
days inactive
5706
days old
8 comments
6 participants
participants (6)
-
Edson Tirelli -
James Owen -
Mark Proctor -
Michael Neale -
mrhoden＠franklinamerican.com -
Thomas Frenken