TY - JOUR
T1 - A Survey of GPU Multitasking Methods Supported by Hardware Architecture
AU - Zhao, Chen
AU - Gao, Wu
AU - Nie, Feiping
AU - Zhou, Huiyang
N1 - Publisher Copyright:
© 2022 IEEE Computer Society. All rights reserved.
PY - 2022/6/1
Y1 - 2022/6/1
N2 - The ability to support multitasking becomes more and more important in the development of graphic processing unit (GPU). GPU multitasking methods are classified into three types: temporal multitasking, spatial multitasking, and simultaneous multitasking (SMK). This article first introduces the features of some commercial GPU architectures to support multitasking and the common metrics used for evaluating the performance of GPU multitasking methods, and then reviews the GPU multitasking methods supported by hardware architecture (i.e., hardware GPU multitasking methods). The main problems of each type of hardware GPU multitasking methods to be solved are illustrated. Meanwhile, the key idea of each previous hardware GPU multitasking method is introduced. In addition, the characteristics of hardware GPU multitasking methods belonging to the same type are compared. This article also gives some valuable suggestions for the future research. An enhanced GPU simulator is needed to bridge the gap between academia and industry. In addition, it is promising to expand the research space with machine learning technologies, advanced GPU architectural innovations, 3D stacked memory, etc. Because most previous GPU multitasking methods are based on NVIDIA GPUs, this article focuses on NVIDIA GPU architecture, and uses NVIDIA’s terminology. To our knowledge, this article is the first survey about hardware GPU multitasking methods. We believe that our survey can help the readers gain insights into the research field of hardware GPU multitasking methods.
AB - The ability to support multitasking becomes more and more important in the development of graphic processing unit (GPU). GPU multitasking methods are classified into three types: temporal multitasking, spatial multitasking, and simultaneous multitasking (SMK). This article first introduces the features of some commercial GPU architectures to support multitasking and the common metrics used for evaluating the performance of GPU multitasking methods, and then reviews the GPU multitasking methods supported by hardware architecture (i.e., hardware GPU multitasking methods). The main problems of each type of hardware GPU multitasking methods to be solved are illustrated. Meanwhile, the key idea of each previous hardware GPU multitasking method is introduced. In addition, the characteristics of hardware GPU multitasking methods belonging to the same type are compared. This article also gives some valuable suggestions for the future research. An enhanced GPU simulator is needed to bridge the gap between academia and industry. In addition, it is promising to expand the research space with machine learning technologies, advanced GPU architectural innovations, 3D stacked memory, etc. Because most previous GPU multitasking methods are based on NVIDIA GPUs, this article focuses on NVIDIA GPU architecture, and uses NVIDIA’s terminology. To our knowledge, this article is the first survey about hardware GPU multitasking methods. We believe that our survey can help the readers gain insights into the research field of hardware GPU multitasking methods.
KW - GPU multitasking
KW - hardware architecture
KW - simultaneous multitasking (SMK)
KW - spatial multitasking
KW - survey
KW - temporal multitasking
UR - http://www.scopus.com/inward/record.url?scp=85116882411&partnerID=8YFLogxK
U2 - 10.1109/TPDS.2021.3115630
DO - 10.1109/TPDS.2021.3115630
M3 - 文章
AN - SCOPUS:85116882411
SN - 1045-9219
VL - 33
SP - 1451
EP - 1463
JO - IEEE Transactions on Parallel and Distributed Systems
JF - IEEE Transactions on Parallel and Distributed Systems
IS - 6
ER -