计算机工程与应用2016，Vol.52Issue(21)：30-35,6.DOI:10.3778/j.issn.1002-8331.1603-0434
面向CPU-GPU源到源编译系统的渐近拟合优化方法
Asymptotic fitting optimization technology for source-to-source compile system on CPU-GPU architecture
摘要
Abstract
Aiming at addressing the problem of the inadequate performance optimization after developing and porting of application on CPU-GPU heterogeneous parallel system, a new approach for CPU-GPU system is proposed, which com-bines asymptotic fitting optimization with source-to-source compiling technique. This approach can translate C code that inserts directives into CUDA code, and profile the generated code several times. Besides, the approach can realize the source-to-source compiling and optimization of the generated code automatically, and a prototype system based on the approach is realized in this paper as well. Functionality and performance evaluations of the prototype show that the gener-ated CUDA code is functionally equivalent to the original C code while its improvement in performance is significant. When compared with CUDA benchmark, the performance of the generated CUDA code is obviously better than codes generated by other source-to-source compiling technique.关键词
源到源编译/统一计算架构(CUDA)/剖分/渐近拟合优化Key words
source-to-source compile/Compute Unified Device Architecture(CUDA)/profiling/asymptotic fitting optimization分类
信息技术与安全科学引用本文复制引用
魏洪昌,朱正东,董小社,宁洁..面向CPU-GPU源到源编译系统的渐近拟合优化方法[J].计算机工程与应用,2016,52(21):30-35,6.基金项目
国家自然科学基金（No.61572394）；国家高技术研究发展计划（863）（No.2014AA01A302）。 （）
