摘要
由于GPU可以加速深度学习作业的处理,许多研究人员通过提高GPU利用率来达到减少作业完成时间的目的。与传统的作业独占GPU资源来减少作业完成时间不同,考虑了多个作业共置的问题(即同一个GPU中同时执行多个作业能有效提高GPU利用率并减少作业完成时间),提出了一种面向深度学习作业的干扰感知在线调度算法(OASIS)。该算法首先在作业共置的情况下,使用改进的机器学习方法构建了作业所需资源的预测模型。其次,为了计算作业间干扰值,设计了一种作业组合模型,通过模型计算的干扰值来主动修改作业调度策略以避免无效调度,达到减少作业完成时间的目的。最后,在真实环境中部署了实验,实验结果表明:提出的OASIS算法与经典的FCFS算法、MBP算法和SJF算法相比,不仅平均作业总体完成时间缩短了5.7%,而且平均能耗降低了4.0%,验证结果充分说明了该算法的有效性和优越性。
Since GPU can accelerate the processing of deep learning jobs,many researchers aim to reduce job completion time by improving GPU utilization.Different from the traditional approach of dedicating GPU resources to a single job to reduce completion time,this paper considers the issue of job colocation(i.e.,executing multiple jobs simultaneously on the same GPU to effectively improve GPU utilization and reduce job completion time)and proposes an interference-aware online scheduling algorithm for deep learning jobs(OASIS).This algorithm first uses an improved machine learning approach to construct a prediction model for the resources required by jobs in the context of job colocation.Then,to calculate the interference values between jobs,a job combination model is designed.The interference values calculated by this model are used to proactively adjust the job scheduling strategy to avoid ineffective scheduling,thereby reducing job completion time.Finally,experiments are deployed in a real-world environment,and the results show that compared to the classical FCFS,MBP,and SJF algorithms,the proposed OASIS algorithm not only reduces the average total job completion time by 5.7%,but also decreases the average energy consumption by 4.0%.These results fully demonstrate the effectiveness and superiority of the proposed algorithm.
作者
敬超
闭玉申
JING Chao;BI Yu-shen(College of Computer Science and Engineering,Guilin University of Technology,Guilin 541006;Guangxi Key Laboratory of Embedded Technology and Intelligent System,Guilin University of Technology,Guilin 541006,China)
出处
《计算机工程与科学》
CSCD
北大核心
2024年第12期2138-2148,共11页
Computer Engineering & Science
基金
国家自然科学基金(62362018)。
关键词
深度学习
干扰感知
资源预测模型
在线调度
deep learning
interference-aware
resource prediction model
online scheduling
作者简介
敬超(1983-),男,河南长葛人,博士,副教授,CCF会员(41171M),研究方向为高性能计算、节能调度和异构计算系统上的功耗预测。E-mail:jingchao@glut.edu.cn;闭玉申(1997-),男,广西南宁人,硕士生,研究方向为人工神经网络、功耗预测和异构计算系统的调度。E-mail:biyushen@glut.edu.cn。
