However, the scheduling problem is over simplified that treats the compute cluster as a single collection of resources, which is unpractical in real systems. The policy gradient based REINFORCE algorithm is used to train a fully connected policy network with 20 neurons. proposed using deep reinforcement learning to solve a simplified task s scheduling problem. However, the job-shop scheduling is different from the DAG tasks scheduling problem, where DAG tasks have more complex dependencies and data communication cost. first proposed using classic reinforcement learning to address job-shop scheduling problem. Experimental results showed the effectiveness of the proposed algorithm compared with the classic HEFT/CPOP algorithms. Leveraging the algorithm’s capability of exploring long term reward, the proposed algorithm could achieve good scheduling policies while guaranteeing trusted tasks scheduled within trusted entities. The MCTS method is proposed to determine actual scheduling policies when DAG tasks are simultaneously executed in trusted and untrusted entities. Efficient scheduling state space, action space and reward function are designed to train the policy gradient-based REINFORCE agent. The scheduling problem is defined using the reinforcement learning model. In this paper, we propose a trust-aware adaptive DAG tasks scheduling algorithm using the reinforcement learning and Monte Carlo Tree Search (MCTS) methods. However, many previously proposed traditional heuristic algorithms are usually based on greedy methods and also lack the consideration of scheduling tasks between trusted and untrusted entities, which makes the problem more complicated, but there still exists a large optimization space to be explored. Due to its key importance, the DAG tasks scheduling problem has been extensively studied in the literature. ![]() The Directed Acyclic Graph (DAG) tasks scheduling problem is NP-complete and it is hard to find an optimal schedule. Task scheduling is critical for improving system performance in the distributed heterogeneous computing environment.
0 Comments
Leave a Reply. |