Hi Everybody, This may be a redundant question and may have been addressed or mentioned somewhere. Is there a rule of thumb as to when a constraint should be defined -- before distribution or after distribution? There are certain constraint definition for our problem, it seems it is easier to define them after distribution (i.e., after a set of resources have been chosen for tasks). Once the space fails for any constraint definition it goes back and selects a different set of resources and so on. By putting the constraint before the distribution that we have done for some our constraints, we think it reduces the computation time. But does it increase the memory consumption? Now that we are getting a little more knowledgeable in Mozart/Oz we think we can define these constraints either before or after the distribution. However, the performance that we are seeing, seems not to fit always what we were hoping for! But on balance does putting the constraints before distribution consume more memory? How about computation time? Are there any guidelines for the best practice?
Thanks, Ashis _________________________________________________________________________________ mozart-users mailing list [email protected] http://www.mozart-oz.org/mailman/listinfo/mozart-users
