基于模块化微服务的GPU利用率管理器

Arun A. Balakrishnan, S. Kamal, C. SatheeshChandran
{"title":"基于模块化微服务的GPU利用率管理器","authors":"Arun A. Balakrishnan, S. Kamal, C. SatheeshChandran","doi":"10.46243/jst.2020.v5.i4.pp230-237","DOIUrl":null,"url":null,"abstract":":Graphics processing unit (GPU) is a computer programmable chip that could perform rapid\nmathematical operations that can be accelerated with massive parallelism. In the early days, central processing unit\n(CPU) was responsible for all computations irrespective of whether it is feasible for parallel computation. However,\nin recent years GPUs are increasingly used for massively parallel computing applications, such as training Deep\nNeural Networks. GPU’s performance monitoring plays a key role in this new era since GPUs serve an inevitable\nrole in increasing the speed of analysis of the developed system. GPU administration comes in picture to efficiently\nutilize the GPU when we deal with multiple workloads to run on the same hardware. In this study, various GPUparameters are monitored and help to keep them in safe levels and also to keep the improved performance of the\nsystem. This study,","PeriodicalId":23534,"journal":{"name":"Volume 5, Issue 4","volume":"83 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2020-07-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Modular Microservice based GPU Utilization Manager with\\nGunicorn\",\"authors\":\"Arun A. Balakrishnan, S. Kamal, C. SatheeshChandran\",\"doi\":\"10.46243/jst.2020.v5.i4.pp230-237\",\"DOIUrl\":null,\"url\":null,\"abstract\":\":Graphics processing unit (GPU) is a computer programmable chip that could perform rapid\\nmathematical operations that can be accelerated with massive parallelism. In the early days, central processing unit\\n(CPU) was responsible for all computations irrespective of whether it is feasible for parallel computation. However,\\nin recent years GPUs are increasingly used for massively parallel computing applications, such as training Deep\\nNeural Networks. GPU’s performance monitoring plays a key role in this new era since GPUs serve an inevitable\\nrole in increasing the speed of analysis of the developed system. GPU administration comes in picture to efficiently\\nutilize the GPU when we deal with multiple workloads to run on the same hardware. In this study, various GPUparameters are monitored and help to keep them in safe levels and also to keep the improved performance of the\\nsystem. This study,\",\"PeriodicalId\":23534,\"journal\":{\"name\":\"Volume 5, Issue 4\",\"volume\":\"83 1\",\"pages\":\"\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2020-07-30\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Volume 5, Issue 4\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.46243/jst.2020.v5.i4.pp230-237\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Volume 5, Issue 4","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.46243/jst.2020.v5.i4.pp230-237","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

摘要

图形处理单元(GPU)是一种计算机可编程芯片,可以执行快速的数学运算,可以通过大规模并行加速。在早期,中央处理器(CPU)负责所有的计算,而不管它是否适合并行计算。然而,近年来gpu越来越多地用于大规模并行计算应用,例如训练深度神经网络。GPU的性能监控在这个新时代起着至关重要的作用,因为GPU在提高开发系统的分析速度方面发挥着不可避免的作用。当我们在同一硬件上处理多个工作负载时,GPU管理可以有效地利用GPU。在本研究中,对各种gpu参数进行监测,以帮助将它们保持在安全水平,并保持系统性能的提高。这项研究中,
本文章由计算机程序翻译,如有差异,请以英文原文为准。
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
Modular Microservice based GPU Utilization Manager with Gunicorn
:Graphics processing unit (GPU) is a computer programmable chip that could perform rapid mathematical operations that can be accelerated with massive parallelism. In the early days, central processing unit (CPU) was responsible for all computations irrespective of whether it is feasible for parallel computation. However, in recent years GPUs are increasingly used for massively parallel computing applications, such as training Deep Neural Networks. GPU’s performance monitoring plays a key role in this new era since GPUs serve an inevitable role in increasing the speed of analysis of the developed system. GPU administration comes in picture to efficiently utilize the GPU when we deal with multiple workloads to run on the same hardware. In this study, various GPUparameters are monitored and help to keep them in safe levels and also to keep the improved performance of the system. This study,
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
自引率
0.00%
发文量
0
期刊最新文献
Picasso’nun mavi dönem resimlerinde melankoli kavramının yeri Niğde Müzesi teşhir salonu ve deposunda bulunan halı örnekleri Yeni Medya platformlarında sanal gerçeklik uygulamalarının geleceği ve bilim kurgu evrenindeki yansımaları Effect of The Covid-19 Pandemic Period on Zero Waste Awareness: A Scale Development Survey Rembrandt’ın resimlerinde Doğu dünyasına ait unsurların sanatsal açıdan incelenmesi
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1