Hmmmm.... With today's 64-bit environments, you can do several Terabytes in memory; if you have the memory and the architecture. We did some studies on that back in 2002 and found that 4GB was terribly limiting to a "really complex" rulebase. 100 tables with 1,000,000 entries ends up being 100,000,000 entries which would take, on the average of 2,000 bytes per object, 200,000,000,000 bytes of memory. And that's BEFORE you add the complexity of the rule matching. That's quite a bit and smacks of poor architecture. It's the old principle of "Just throw everything into a big bucket, stir it around, and see what rises to the top." No thought. No planning. Just hoping.

Now, to the time problem; I think Dr. Friedman-Hill has covered this many times and it's covered quite well in his book, "Jess in Action" - available from Amazon. And, no, I'm not pimping for the book. Just pointing it out.


Performance tests are available at http://www.kbsc.com/ benchmarks2000-2005.xls (available sometime tonight or tomorrow) and (soon - maybe a couple of months) http://www.kbsc.com/ benchmarks2006-2010.xls. Enjoy...

SDG
jco

James C. Owen
Senior Consultant
[EMAIL PROTECTED]
http://www.kbsc.com
"Never give up.  Never give up.  Never, never, never give up."
From a speech by former Prime Minister, Winston Churchill, during the most trying times of the British Empire; a great leader too soon forgotten by his country after keeping it from being swallowed up by the Nazi Empire during the second world war.



On Feb 7, 2006, at 11:18 AM, Thomas Beer wrote:

Hi all Jess users!
I'm not sure if this mailing list is the right one for my question, if there
is a more adequate list just give me a hint.
I have a more common question regarding Jess, rule-based expert systems and deductive databases. What are the essential differences (besides the smooth
Java integration) between Jess (or an rule-based expert system) and
deductive databases (e.g. Ontobroker)?
Is it possible to store a huge amount of facts (related to databases, e.g. 100 tables with 1000000 table entries each) in the working memory? Is there
a kind of database backend integrated which automatically updates the
working memory? Are there some performance tests available?
Is Jess appropriate for handling "time sensitive" facts and rules related to them, e.g. "If time is 06:00 then ring alarm". What I'm asking is, is it possible (or reasonable) using Jess to implement time dependent triggers? This would require periodically updates (e.g. every milli second) of the
working memory, right?!?

I strongly appreciate your help!

Best regards,
Thomas Beer



Thomas Beer
ECCA - eTourism Competence Center Austria
Technikerstrasse 21a
ICT-Technologiepark
6020 Innsbruck
Austria

E-Mail: [EMAIL PROTECTED]
www.etourism-austria.at

--------------------------------------------------------------------
To unsubscribe, send the words 'unsubscribe jess-users [EMAIL PROTECTED]'
in the BODY of a message to [EMAIL PROTECTED], NOT to the list
(use your own address!) List problems? Notify owner-jess- [EMAIL PROTECTED]
--------------------------------------------------------------------


--------------------------------------------------------------------
To unsubscribe, send the words 'unsubscribe jess-users [EMAIL PROTECTED]'
in the BODY of a message to [EMAIL PROTECTED], NOT to the list
(use your own address!) List problems? Notify [EMAIL PROTECTED]
--------------------------------------------------------------------

Reply via email to