It can. Good point.
That's multi-threading or multi-vm'ing the solver algorithm, not the
problem itself.
The second to last paragraph briefly mentions that too IIRC.
OptaPlanner will support this (hopefully for 6.2)
without breaking incremental score calculation:
https://issues.jboss.org/browse/PLANNER-76
The difficulty is not breaking incremental score calculation,
which performance loss is far greater than multi-threading/vm can gain.
On 12-03-14 07:50, john poole wrote:
I was reading the blog
http://www.optaplanner.org/blog/2014/03/03/CanMapReduceSolvePlanningProbl...
and although the unsuitability of the spatial division of the problem makes
sense, I'd like to understand why the problem can't be broken up
differently. i.e. why can't any solution be loaded into n separate solvers
with 1/nth of the moves, and then send the best solution to all the solvers,
and repeat?
--
View this message in context:
http://drools.46999.n3.nabble.com/Non-Spatial-Map-Reduce-tp4028652.html
Sent from the Drools: User forum mailing list archive at
Nabble.com.
_______________________________________________
rules-users mailing list
rules-users(a)lists.jboss.org
https://lists.jboss.org/mailman/listinfo/rules-users