Skip to Main Content
The advantages of integrating reinforcement learning (RL) techniques into scientific parallel time-stepping applications have been revealed in research work over the past few years. The object of the integration was to automatically select the most appropriate dynamic loop scheduling (DLS) algorithm from a set of available algorithms with the purpose of improving the application performance via load balancing during the application execution. This paper investigates the performance of such a dynamic loop scheduling with reinforcement learning (DLS-with-RL) approach to load balancing. The DLS-with-RL is most suitable for use in time-stepping scientific applications with large number of steps. The RL agent's characteristics depend on a learning rate parameter and a discount factor parameter. An application simulating wavepacket dynamics that incorporates a DLS-with-RL approach is allowed to execute on a cluster of workstations to investigate the influences of these parameters. The RL agent implemented two RL algorithms: QLEARN and SARSA learning. Preliminary results indicate that on a fixed number of processors, the simulation completion time is not sensitive to the values of the learning parameters used in the experiments. The results also indicate that for this application, there is no advantage of choosing one RL technique over another, even though the techniques differed significantly in the number of times they selected the various DLS algorithms.