MCTS Playouts Parallelization with a MPPA Architecture
Résumé
We present a study of the use of a MultiPurpose Processor Array (MPPA) architecture for the parallelization of MCTS algorithms applied to the field of General Game Playing. We evaluate the constraints imposed by this architecture and show that the only feasible MCTS parallelization on MPPA is a leaf parallelization. We show that the MPPA provides good scalability when increasing the size of the communications which is useful when using synchronous communications to send large sets of game initial positions to be processed. We consider two approaches for the calculation of playouts: the distributed computing of a playout on each cluster and the calculation of several playouts per cluster; the second approach gives better results. Finally, we describe experiments concerning the thread management and present a surprising result: it is more efficient to create new threads than to synchronize permanent threads.
Origine : Fichiers produits par l'(les) auteur(s)
Loading...