Deep Reinforcement Agent for Scheduling in HPC
Authors:
- 1 May 2021
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 807-816
Abstract
Cluster scheduler is crucial in high-performance computing (HPC). It determines when and which user jobs should be allocated to available system resources. Existing cluster scheduling heuristics are developed by human experts based on their experience with specific HPC systems and workloads. However, the increasing complexity of computing systems and the highly dynamic nature of application workloads have placed tremendous burden on manually designed and tuned scheduling heuristics. More aggressive optimization and automation are needed for cluster scheduling in HPC. In this work, we present an automated HPC scheduling agent named DRAS (Deep Reinforcement Agent for Scheduling) by leveraging deep reinforcement learning. DRAS is built on a novel, hierarchical neural network incorporating special HPC scheduling features such as resource reservation and backfilling. A unique training strategy is presented to enable DRAS to rapidly learn the target environment. Once being provided a specific scheduling objective given by system manager, DRAS automatically learns to improve its policy through interaction with the scheduling environment and dynamically adjusts its policy as workload changes. The experiments with different production workloads demonstrate that DRAS outperforms the existing heuristic and optimization approaches by up to 45%.Keywords
All Related Versions
Funding Information
- National Science Foundation
- Office of Science
- National Energy Research Scientific Computing Center
- Office of Science
This publication has 16 references indexed in Scilit:
- Mastering the game of Go without human knowledgeNature, 2017
- Trade-Off Between Prediction Accuracy and Underestimation Rate in Job Runtime Estimates2017
- Deep Reinforcement Learning framework for Autonomous DrivingElectronic Imaging, 2017
- Resource Management with Deep Reinforcement Learning2016
- Deep learningNature, 2015
- Energy-efficient and thermal-aware resource management for heterogeneous datacentersSustainable Computing: Informatics and Systems, 2014
- Multi-resource packing for cluster schedulers2014
- Integrating dynamic pricing of electricity into energy aware scheduling for HPC systems2013
- Self-Optimizing Memory Controllers: A Reinforcement Learning Approach2008
- Utilization, predictability, workloads, and user runtime estimates in scheduling the IBM SP2 with backfillingIEEE Transactions on Parallel and Distributed Systems, 2001