A HPC Co-Scheduler with Reinforcement Learning
Preprint
Abstract
Although High Performance Computing (HPC) users understand basic resource requirements such as the number of CPUs and memory limits, internal infrastructural utilization data is exclusively leveraged by cluster operators, who use it to configure batch schedulers. This task is challenging and increasingly complex due to ever larger cluster scales and heterogeneity of modern scientific workflows. As a result, HPC systems achieve low utilization with long job completion times (makespans). To tackle these challenges, we propose a co-scheduling algorithm based on an adaptive reinforcement learning algorithm, where application profiling is combined with cluster monitoring. The resulting cluster scheduler matches resource utilization to application performance in a fine-grained manner (i.e., operating system level). As opposed to nominal allocations, we apply decision trees to model applications' actual resource usage, which are used to estimate how much resource capacity from one allocation can be co-allocated to additional applications. Our algorithm learns from incorrect co-scheduling decisions and adapts from changing environment conditions, and evaluates when such changes cause resource contention that impacts quality of service metrics such as jobs slowdowns. We integrate our algorithm in an HPC resource manager that combines Slurm and Mesos for job scheduling and co-allocation, respectively. Our experimental evaluation performed in a dedicated cluster executing a mix of four real different scientific workflows demonstrates improvements on cluster utilization of up to 51% even in high load scenarios, with 55% average queue makespan reductions under low loads.Keywords
All Related Versions
- Version 1, 2024-01-18, ArXiv
- Published version: , 126.
This publication has 10 references indexed in Scilit:
- Dynamic Colocation Policies with Reinforcement LearningACM Transactions on Architecture and Code Optimization, 2020
- Learning scheduling algorithms for data processing clusters2019
- Speculative Scheduling for Stochastic HPC Applications2019
- PMIx: Process management for exascale environmentsParallel Computing, 2018
- Scalable system scheduling for HPC and big dataJournal of Parallel and Distributed Computing, 2018
- SLO-aware colocation of data center tasks based on instantaneous processor requirements2017
- Experience with using the Parallel Workloads ArchiveJournal of Parallel and Distributed Computing, 2014
- The NIST definition of cloud computing2011
- Gapped BLAST and PSI-BLAST: a new generation of protein database search programsNucleic Acids Research, 1997
- State of the Art—A Survey of Partially Observable Markov Decision Processes: Theory, Models, and AlgorithmsManagement Science, 1982