Optimizing HPC scheduling: a hierarchical reinforcement learning approach for intelligent job selection and allocation
Open Access
- 29 May 2025
- journal article
- research article
- Published by Springer Nature in The Journal of Supercomputing
- Vol. 81 (8) , 1-32
Abstract
High-performance computing (HPC) systems face increasing challenges in job scheduling due to the evolving complexity of computational tasks and the growing diversity and heterogeneity of resources. Job allocation is a critical aspect in contemporary HPC systems, due to compute nodes possessing an increased capacity in terms of physical resources and having the capability to execute multiple jobs simultaneously. However, job allocation is often overlooked in existing reinforcement learning (RL)-based schedulers that mainly focus on selecting suitable jobs from the job queue and leave allocation to overly simplistic policies, such as first-available allocation. The bin-packing nature at the node level of modern HPC necessitates more refined and intelligent allocation strategies. This paper introduces HeraSched, a novel hierarchical reinforcement learning (HRL)-based scheduler, adept at intelligent job selection without separate backfilling and heterogeneity-aware allocation, tailored for modern HPC environments. It efficiently manages diverse workloads across CPU and GPU cluster partitions. We evaluate HeraSched using real-world workloads, demonstrating significant improvements in reducing job waiting times and preventing job starvation compared to 27 scheduling combinations. In validation, the best maximum waiting time among compared methods is 78% higher than HeraSched’s result in overloaded CPU partitions. This performance demonstrates HeraSched’s ability to manage intensely stressed workloads and adapt to previously unseen, high-demand scenarios, thereby establishing a new standard in HPC job scheduling.Keywords
Funding Information
- University of Melbourne
This publication has 40 references indexed in Scilit:
- Optimization Metrics for the Evaluation of Batch Schedulers in HPC2023
- An SMDP approach for Reinforcement Learning in HPC cluster schedulersFuture Generation Computer Systems, 2022
- GARLSched: Generative adversarial deep reinforcement learning task scheduling optimization for large-scale high performance computing systemsFuture Generation Computer Systems, 2022
- Modeling and simulation of hierarchical task allocation system for energy-aware HPC cloudsSimulation Modelling Practice and Theory, 2021
- Efficient task allocation approach using genetic algorithm for cloud environmentCluster Computing, 2019
- Experience with using the Parallel Workloads ArchiveJournal of Parallel and Distributed Computing, 2014
- Heuristics for Energy-Aware VM Allocation in HPC Clouds2014
- A survey on resource allocation in high performance distributed computing systemsParallel Computing, 2013
- Computational models and heuristic methods for Grid scheduling problemsFuture Generation Computer Systems, 2010
- Reinforcement Learning1992