I wanted to find out if anyone has a comment on what would increase the performance of the rules more. 1. Removing the typecasting from strings to enumerations back to strings for 1/3 of all the rules 2. Splitting a rule that gets executed for every request multiple times with 4 accumulates with sum into 4 different rules with single accumulates with sum My guess is that the second one would increase the performance more because keeping all the accumulates in one rule, every time the first accumulate changes, the other accumulates would have to be recalculated. This could exponentially affect the performance. Any comments on this would be greatly appreciated. -- View this message in context: http://drools.46999.n3.nabble.com/Performance-boost-separating-accumulate... Sent from the Drools: User forum mailing list archive at Nabble.com.