[rules-users] high memory usage for rules with more than 6 facts as condition

[Wolfgang Laun]

I was assuming that repeated matches of the same element amon the 10 are
excluded (which they probably aren't), so Thomas' formula is the correct one.
-W

On Wed, Apr 14, 2010 at 12:19 PM, Wolfgang Laun <wolfgang.laun at gmail.com> wrote:
> Do you constrain the number of possible matches of 10 elements, which,
> without constraints, is quite a number (10! = 3628800)?
>
> Tests that produce a highly unlikely scenario aren't particularly useful.
>
> -W
>
> On Wed, Apr 14, 2010 at 11:40 AM, murphy <christian.niehues at its-telco.de> wrote:
>>
>> Hello,
>>
>> I did some tests to get an opinion about the performance of drools (5.0.1).
>>
>> At one test I have a rule with 10 objects (same object-type) declared as
>> condition and got an OutOfMemoryException. I noticed that the memory usage
>> increases when I insert the last facts into my
>> StatefulKnowledgeSession. On fact 7 it increases with 10MB, on fact 8 with
>> 46MB and on fact 9 with 170MB.
>>
>> I debuged a little bit in the drools-source and noticed that the
>> LeftTupleMemory increased exponential with the number of facts I inserted.
>>
>> My question about this behaviour:
>> Is a condition limited to a handful of facts or do I have to change the
>> syntax to get it working? I know that I could insert the same object-types
>> as a list, but assume that I have 10 facts with different object-types.
>>
>> Thanks in reply,
>> Christian
>>
>> --
>> View this message in context: http://n3.nabble.com/high-memory-usage-for-rules-with-more-than-6-facts-as-condition-tp718114p718114.html
>> Sent from the Drools - User mailing list archive at Nabble.com.
>> _______________________________________________
>> rules-users mailing list
>> rules-users at lists.jboss.org
>> https://lists.jboss.org/mailman/listinfo/rules-users
>>
>