Matrix multiplication implementation based on array types of Parray

doi:10.3724/SP.J.1016.2014.02564

CORC > 北京大学 > 信息科学技术学院

	Matrix multiplication implementation based on array types of Parray
	Cui, Xiang ; Li, Xiao-Wen ; Chen, Yi-Feng
刊名	jisuanji xuebaochinese journal of computers
	2014
DOI	10.3724/SP.J.1016.2014.02564
英文摘要	In this paper, a programming interface of GPU-accelerated heterogeneous clusters named Parray is introduced. In Parray, the concept of array type is involved to separate the physical data layout and logical structure of multi-dimensional data; the control flow diversion of heterogeneous computation units is formally unified. An example code of matrix multiplication is shown to demonstrate the programming characteristics of Parray: the code envolves from a single CPU-thread code to multi-threads code and then a GPU code by modifying the array types and several program lines. A GPU-based high performance GEMM implemented in Parray is introduced and achieves almost the same Gflops when testing on a single node of Tian-1A system. Because the code operates directly on the logical structure of data, the same GEMM code can work on different physical array data layouts.; EI; 0; 12; 2563-2573; 37
语种	英语
内容类型	期刊论文
源URL	[http://ir.pku.edu.cn/handle/20.500.11897/327805]
专题	信息科学技术学院
推荐引用方式 GB/T 7714	Cui, Xiang,Li, Xiao-Wen,Chen, Yi-Feng. Matrix multiplication implementation based on array types of Parray[J]. jisuanji xuebaochinese journal of computers,2014.
APA	Cui, Xiang,Li, Xiao-Wen,&Chen, Yi-Feng.(2014).Matrix multiplication implementation based on array types of Parray.jisuanji xuebaochinese journal of computers.
MLA	Cui, Xiang,et al."Matrix multiplication implementation based on array types of Parray".jisuanji xuebaochinese journal of computers (2014).