A user mode CPU–GPU scheduling framework for hybrid workloads

Wang, Bin; Ma, Ruhui; Qi, Zhengwei; Yao, Jianguo; Guan, Haibing

doi:10.1016/j.future.2016.03.011

Cited by 9 publications

(1 citation statement)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Data classification storage is a very important task of memory, but also a very tricky and heavy task. According to the data type, we can make data analysis and sorting classification, and according to the appropriate criteria for matching [11]. When the data classification is completed, the next will be more arduous work, that is, to find the most suitable for storing each type of data storage space, and then the data movement.…”

Section: Storage Technology Optimization For Cudamentioning

confidence: 99%

Research on Programming Model and Compilation Optimization Technology of Multi-Core GPU

Chen

2022

J. Phys.: Conf. Ser.

View full text Add to dashboard Cite

GPGPU (General Purpose Computing on Graphics Processing Units) has been widely applied to high performance computing. However, GPU architecture and programming model are different from that of traditional CPU. Accordingly, it is rather challenging to develop efficient GPU applications. This paper focuses on the key techniques of programming model and compiler optimization for many-core GPU, and addresses a number of key theoretical and technical issues. This paper proposes a many-threaded programming model ab-Stream, which would transparentize architecture differences and provide an easy to parallel, easy to program, easy to extend and easy to tune programming model. In addition, this paper proposes memory optimization and data transfer transformation according to data classification. Firstly, this paper proposes data layout pruning based on classification memory, and then proposes Ta T (Transfer after Transformed) for transferring Strided data between CPU and GPU. Experimental results demonstrate that proposed techniques would significantly improve performance for GPGPU applications.

show abstract

Section: Storage Technology Optimization For Cudamentioning

confidence: 99%