At 11:25 29/6/2001 -0700, friedman_hill ernest j wrote:
> Thank you for sharing these very interesting results! This corresponds
> well with what I previously believed: generally, that Forgy's Rete-II
> algorithm was the fastest on this benchmark, that otherwise Jess was
> the fastest of the non-Rete-II Java-based system and within a factor of
> two of the best non-Rete-II native systems.
I agree with you on this.
> -Except- for the "JRULES-OPT" results. Are you sure those are valid?
> What is that product, exactly? I find it very hard to believe that
> anything could beat OPSJ on this task by a factor of 4.
Yes, those results are very much valid! JRULES-OPT stands for JRules
in its optimized mode. In this mode, JRules makes use of auto-hashing
and finders to improve the performance of the engine. The finders
feature requires overwriting certain methods in certain classes
(although not the rules) and must be done on a per-application basis.
In this manners test, the optimization improves dramatically the
performance. However, in other tests, it was simply irrelevant.
Regards,
Juanjo
--
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~+
| Juan José García Adeva |
| Room 1G-628A mailto:[EMAIL PROTECTED] |
| Avaya Labs Research http://www.avayalabs.com |
| 101 Crawdords Corner Road Phone: +1-732-817-6228 |
| Holmdel, NJ 0733-3030 Fax: +1-732-817-4870 |
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~+
---------------------------------------------------------------------
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]
---------------------------------------------------------------------
