• 전체
  • 전자/전기
  • 통신
  • 컴퓨터
로그인 회원가입 About Us 이용안내
사이트맵
닫기

사이트맵

Loading..

Please wait....

홈 홈 > 연구문헌 >

Current Result Document :

한글제목(Korean Title) 고집적 GPU 서버의 데이터 병렬 딥러닝 성능 및 확장성
영문제목(English Title) Performance and Scalability of Many-GPU Server for Deep Learning Applications
저자(Author) 이혁민   문호상   이재희   김태영   박소윤   오승연   정성택   Hyeok-Min Lee   Ho-Sang Moon   Jae-Heui Lee   Tae-Young Kim So-Yun Bak   Seung-Yeon Oh   Sung-Taek Chung   김재훈   문호상   이재희   윤성욱   이정은   이동재   정성택   Jae-Hoon Kim   Ho-Sang Moon   Jae-Heui Lee   Sung-Wook Yoon   Jung-Eun Lee   Dong-Jae Lee   Sung-Taek Chung   전용성   이상우   성하영   YongSung Jeon   SangWoo Lee   HaYoung Seong   김영우   배유석   Young woo Kim   Yuseok Bae  
원문수록처(Citation) VOL 45 NO. 01 PP. 2072 ~ 2075 (2022. 06)
한글내용
(Korean Abstract)
영문내용
(English Abstract)		 In this paper, we investigate and execute the performance and scalability analysis for deep learning application especially on a highly integrated many-GPU server environment. In general, a conventional GPU server can equip four to ten GPUs in a server chassis. There are many restrictions to hider equipping many GPUs in a server – for example, physical chassis dimension, number of PCIe connections, and etc. In this paper, we integrate the GPU scalability in a server over ten GPUs per server by utilizing proprietary external PCIe expansion hardware, and investigate GPU scale-up performance by applying a deep learning application. The implementation and experimentation result show that a GPU server can equip up to twenty-six GPUs - the total number of GPUs in a server is limited by BIOScapability, and its performance scaled up linearly in a server.
키워드(Keyword)
파일첨부 PDF 다운로드

목록