Highly Optimized Code Generation for Stencil Codes with Computation Reuse for GPUs

CORC > 软件研究所 > 软件所图书馆 > 期刊论文

	Highly Optimized Code Generation for Stencil Codes with Computation Reuse for GPUs
	Ma, WJ ; Gao, K ; Long, GP
刊名	JOURNAL OF COMPUTER SCIENCE AND TECHNOLOGY
	2016
卷号	31 期号:6 页码:1262-1274
关键词	GPGPU OpenCL stencil code generation computation reuse
ISSN号	1000-9000
中文摘要	Computation reuse is known as an effective optimization technique. However, due to the complexity of modern GPU architectures, there is yet not enough understanding regarding the intriguing implications of the interplay of computation reuse and hardware specifics on application performance. In this paper, we propose an automatic code generator for a class of stencil codes with inherent computation reuse on GPUs. For such applications, the proper reuse of intermediate results, combined with careful register and on-chip local memory usage, has profound implications on performance. Current state of the art does not address this problem in depth, partially due to the lack of a good program representation that can expose all potential computation reuse. In this paper, we leverage the computation overlap graph (COG), a simple representation of data dependence and data reuse with "element view", to expose potential reuse opportunities. Using COG, we propose a portable code generation and tuning framework for GPUs. Compared with current state-of-the-art code generators, our experimental results show up to 56.7% performance improvement on modern GPUs such as NVIDIA C2050.
英文摘要	Computation reuse is known as an effective optimization technique. However, due to the complexity of modern GPU architectures, there is yet not enough understanding regarding the intriguing implications of the interplay of computation reuse and hardware specifics on application performance. In this paper, we propose an automatic code generator for a class of stencil codes with inherent computation reuse on GPUs. For such applications, the proper reuse of intermediate results, combined with careful register and on-chip local memory usage, has profound implications on performance. Current state of the art does not address this problem in depth, partially due to the lack of a good program representation that can expose all potential computation reuse. In this paper, we leverage the computation overlap graph (COG), a simple representation of data dependence and data reuse with "element view", to expose potential reuse opportunities. Using COG, we propose a portable code generation and tuning framework for GPUs. Compared with current state-of-the-art code generators, our experimental results show up to 56.7% performance improvement on modern GPUs such as NVIDIA C2050.
收录类别	SCI
语种	英语
WOS记录号	WOS:000387335600015
公开日期	2016-12-09
内容类型	期刊论文
源URL	[http://ir.iscas.ac.cn/handle/311060/17294]
专题	软件研究所_软件所图书馆_期刊论文
推荐引用方式 GB/T 7714	Ma, WJ,Gao, K,Long, GP. Highly Optimized Code Generation for Stencil Codes with Computation Reuse for GPUs[J]. JOURNAL OF COMPUTER SCIENCE AND TECHNOLOGY,2016,31(6):1262-1274.
APA	Ma, WJ,Gao, K,&Long, GP.(2016).Highly Optimized Code Generation for Stencil Codes with Computation Reuse for GPUs.JOURNAL OF COMPUTER SCIENCE AND TECHNOLOGY,31(6),1262-1274.
MLA	Ma, WJ,et al."Highly Optimized Code Generation for Stencil Codes with Computation Reuse for GPUs".JOURNAL OF COMPUTER SCIENCE AND TECHNOLOGY 31.6(2016):1262-1274.