作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2019, Vol. 45 ›› Issue (3): 65-72. doi: 10.19678/j.issn.1000-3428.0049676

• 体系结构与软件技术 • 上一篇    下一篇

基于传感器的GPU集群功耗收集监控系统

李超凡,陈庆奎   

  1. 上海理工大学 光电信息与计算机工程学院,上海 200093
  • 收稿日期:2017-12-12 出版日期:2019-03-15 发布日期:2019-03-15
  • 作者简介:李超凡(1990—),男,硕士研究生,主研方向为GPU集群、并行计算、模式识别;陈庆奎(通信作者),教授、博士、博士生导师。
  • 基金资助:

    国家自然科学基金(61572325,60970012);高等学校博士学科点专项科研博导基金(20113120110008);上海重点科技攻关项目(14511107902,16DZ1203603);上海市工程中心建设项目(GCZX14014);上海智能家居大规模物联共性技术工程中心项目(GCZX14014);上海市一流学科建设项目(XTKX2012);沪江基金研究基地专项(C14001)。

GPU Cluster Power Consumption Collection and Monitoring System Based on Sensor

LI Chaofan,CHEN Qingkui   

  1. School of Optical Electrical and Computer Engineering, University of Shanghai for Science and Technology,Shanghai 200093,China
  • Received:2017-12-12 Online:2019-03-15 Published:2019-03-15

摘要:

图像处理器(GPU)集群因其高性能的特性而被广泛应用,但随着GPU规模的增大,其高功耗问题会降低系统的可靠性。为此,提出一种GPU集群功耗收集系统,并设计基于ZigBee无线传感器网络的GPU集群功耗收集监控网络,同时构建收集通信协议和数据库存储系统,通过运行该系统可有效避免通信冲突。实验结果表明,该监控系统可以精确地测量集群中各个GPU的功耗,系统测量误差和丢包率分别低于1%和0.005%。

关键词: 图像处理器集群, ZigBee无线传感器网络, 通信协议, 数据库存储, 低误差低丢包率

Abstract:

Graphics Processing Unit(GPU) clusters are widely used for their high performance characteristics,but with the enlargement of GPU scale,their high power consumption reduce system reliability.Therefore,a GPU cluster power consumption collection system is proposed,and a GPU cluster power collection and monitoring network based on ZigBee Wireless Sensor Network(WSN) is designed.At the same time,a collection communication protocol and a database storage system are constructed.By running the system,communication conflicts can be effectively avoided.Experimental results show that the monitoring network can accurately collect the power consumption of each GPU in the cluster,and the system measurement error and packet loss rate are less than 1% and 0.005% respectively.

Key words: Graphics Processing Unit(GPU) cluster, ZigBee Wireless Sensor Network(WSN), communication protocol, database storage, low error and low packet loss rate

中图分类号: