中国科学院机构知识库网格
Chinese Academy of Sciences Institutional Repositories Grid
首页 机构 成果 学者
  
  1. CAS IR Grid
  2. 计算技术研究所
  3. 中国科学院计算技术研究所
  4. 中国科学院计算技术研究所期刊论文
  5. 英文
EAIS: Energy-aware adaptive scheduling for CNN inference on high-performance GPUs

文献类型：期刊论文

作者Yao, Chunrong3; Liu, Wantao1; Tang, Weiqing2,3; Hu, Songlin1
刊名FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE
出版日期2022-05-01
卷号130页码:253-268
ISSN号0167-739X
关键词Energy-aware Convolutional neural network (CNN) inference High-performance GPUs Workload scheduling Service-Level-Objective (SLO)
DOI10.1016/j.future.2022.01.004
英文摘要Recently, a large number of convolutional neural network (CNN) inference services have emerged on high-performance Graphic Processing Units (GPUs). However, GPUs are high power consumption units, and the energy consumption increases sharply along with the deployment of deep learning tasks. Although previous studies have considered the latency Service-Level-Objective (SLO) of inference services, they fail to directly take account of the energy consumption. Our investigation shows that coordinating batching and dynamic voltage frequency scaling (DVFS) settings can decrease the energy consumption of CNN inference. But it is affected by (i) larger configuration spaces; (ii) GPUs' underutilization while data are transferred between CPUs and GPUs; (iii) fluctuating workloads. In this paper, we propose EAIS, an energy-aware adaptive scheduling framework that is comprised of a performance model, an asynchronous execution strategy, and an energy-aware scheduler. The performance model provides valid information about the performance characteristics of CNN inference services to shrink the feasible configuration space. The asynchronous execution strategy overlaps data upload and GPU execution to improve the system processing capacity. The energy-aware scheduler adaptively coordinates batching and DVFS according to fluctuating workloads to minimize energy consumption while meeting latency SLO. Our experimental results on NVIDIA Tesla M40 and V100 GPUs show that, compared to the state-of-the-art methods, EAIS decreases the energy consumption by up to 28.02% and improves the system processing capacity by up to 7.22% while meeting latency SLO. Besides, EAIS has been proved to have good versatility under different latency SLO constraints. (C) 2022 Elsevier B.V. All rights reserved.
资助项目National Key Research and Development Program of China[2017YFB1010000]
WOS研究方向Computer Science
语种英语
出版者ELSEVIER
WOS记录号WOS:000819692500020
源URL[http://119.78.100.204/handle/2XEOYT63/19518]  
专题中国科学院计算技术研究所期刊论文_英文
通讯作者Liu, Wantao
作者单位1.Chinese Acad Sci, Inst Informat Engn, Beijing 100093, Peoples R China
2.Chinese Acad Sci, Inst Comp Technol, Beijing 100190, Peoples R China
3.Nanjing Univ Sci & Technol, Sch Comp Sci & Engn, Nanjing 210094, Peoples R China
推荐引用方式
GB/T 7714		 Yao, Chunrong,Liu, Wantao,Tang, Weiqing,et al. EAIS: Energy-aware adaptive scheduling for CNN inference on high-performance GPUs[J]. FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE,2022,130:253-268.
APA Yao, Chunrong,Liu, Wantao,Tang, Weiqing,&Hu, Songlin.(2022).EAIS: Energy-aware adaptive scheduling for CNN inference on high-performance GPUs.FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE,130,253-268.
MLA Yao, Chunrong,et al."EAIS: Energy-aware adaptive scheduling for CNN inference on high-performance GPUs".FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE 130(2022):253-268.

入库方式： OAI收割

来源：计算技术研究所

浏览0
下载0
收藏0
其他版本

除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。