While you are at it stick sumatra behind your interfaces - so then we
have 3 algorithm implementations :)
Mark
Peter Lin wrote:
yeah... I don't see why not. feel free to update the performance test
that's already in the examples directory.
peter
On 4/21/06, Alexander Bagerman <[EMAIL PROTECTED]> wrote:
Peter,
Would it be possible to make tests flexible enough that we can profile
reteoo and leaps at the same time? It 's as simple as getting RuleBase
protected RuleBase getLeapsRuleBase() throws Exception {
return new org.drools.leaps.RuleBaseImpl();
}
protected RuleBase getReteRuleBase() throws Exception {
return new RuleBaseImpl();
}
everything else should be pretty much the same
On 4/21/06, Peter Lin <[EMAIL PROTECTED]> wrote:
http://woolfel.blogspot.com/2006/04/more-drools-benchmark-planned.html
http://woolfel.blogspot.com/2006/04/more-drools3-performance.html
Here are some more performance numbers for drools 3, and ideas for
further
tests
peter
