Multi-Agent Deep Reinforcement Learning-Based Resource Allocation in HPC/AI Converged Cluster

作     者：Jargalsaikhan Narantuya Jun-Sik Shin Sun Park JongWon Kim 

作者机构：Department of CloudKakao Enterprise CorpSeongnam13494Korea AI Graduate SchoolGwangju Institute of Science and Technology(GIST)Gwangju61005Korea 

出 版 物：《Computers, Materials & Continua》 (计算机、材料和连续体（英文）)

年 卷 期：2022年第72卷第9期

页      面：4375-4395页

核心收录：

学科分类：08[工学] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：Artificial Intelligence Graduate School Program Institute of Information and Communications Technology Planning and Evaluation Gwangju Institute of Science and Technology, GIST Institute for Information and Communications Technology Promotion, IITP Ministry of Science and ICT, South Korea, MSIT, (2019-0-01842) 

主　　题：Deep learning HPC/AI converged cluster reinforcement learning 

摘      要：As the complexity of deep learning(DL)networks and training data grows enormously,methods that scale with computation are becoming the future of artificial intelligence(AI)*** this regard,the interplay between machine learning(ML)and high-performance computing(HPC)is an innovative paradigm to speed up the efficiency of AI research and ***,building and operating an HPC/AI converged system require broad knowledge to leverage the latest computing,networking,and storage ***,an HPC-based AI computing environment needs an appropriate resource allocation and monitoring strategy to efficiently utilize the system *** this regard,we introduce a technique for building and operating a high-performance AI-computing environment with the latest ***,an HPC/AI converged system is configured inside Gwangju Institute of Science and Technology(GIST),called GIST AI-X computing cluster,which is built by leveraging the latest Nvidia DGX servers,high-performance storage and networking devices,and various open source ***,it can be a good reference for building a small or middlesized HPC/AI converged system for research and educational *** addition,we propose a resource allocation method for DL jobs to efficiently utilize the computing resources with multi-agent deep reinforcement learning(mDRL).Through extensive simulations and experiments,we validate that the proposed mDRL algorithm can help the HPC/AI converged cluster to achieve both system utilization and power consumption *** deploying the proposed resource allocation method to the system,total job completion time is reduced by around 20%and inefficient power consumption is reduced by around 40%.