High performance computer (HPC) is an important tool and means to promote the industrial development. This paper points out several key problems in the development of HPC, and expounds the development trend that the processors are in a transition to a new generation of multi-core heterogeneous parallel computing system, in which the core cache design and multicore heterogeneous programming model design are the key factors. The cloud computing technology may be an important factor to promote the future HPC. Finally, to meet the needs for computers in petroleum exploration, the important role of HPC is further clarified, with the CPU+GPU heterogeneous system as the example. It is pointed out that the development platform and programming method of parallel algorithm are the main bottlenecks of HPC application in petroleum exploration. I/O features of disk, high-speed network and parallel file system are also important factors to affect the HPC application.
LI Min
. Development of High Performance Computer and Its Application to Petroleum Exploration[J]. Science & Technology Review, 2014
, 32(2)
: 80
-83
.
DOI: 10.3981/j.issn.1000-7857.2014.2.013
[1] 田荣, 王迎瑞. 异构架构的新型高效协同模式[C]// 中国高性能计算 2013年会文集, 2013: 751-757. Tian Rong, Wang Yingrui. A new co-processing model on emerging heterogeneous architectures[C]// China High Performance Computing Annual Meeting, 2013: 751-757.
[2] 吴杰谦, 严然, 栾钟治, 等. 云计算环境下资源动态分配方法研究[C]// 中国高性能计算2013年会文集, 2013: 677-680. Wu Jieqian, Yan Ran, Luan Zhongzhi, et al. Research on dynamic re-source allocation in cloud[C]// China High Performance Computing An-nual Meeting, 2013: 677-680.
[3] 刘谷. 可重构众核流处理器上的编译与程序优化技术[D]. 合肥: 中国 科学技术大学, 2013. Liu Gu. Compiler and program optimization on reconfigurable many-core stream processor[D]. Hefei: University of Science and Technology of China, 2013.
[4] 李波. 基于异构多核平台的优化编程研究[D]. 武汉: 华中科技大学, 2011. Li Bo. Research on optimized programming for heterogeneous multicore platform[D]. Wuhan: Huazhong University of Science and Technol-ogy, 2011.
[5] 林一松. 面向GPU的低功耗软件优化关键技术研究[D]. 长沙: 国防科 学技术大学, 2012. Lin Yisong. Study of low power software optimization technology for GPUs[D]. Changsha: National University of Defense Technology, 2012.
[6] 郭御风. 面向多核微处理器芯片的高效能I/O体系结构及其实现技术[D]. 长沙: 国防科学技术大学, 2010. Guo Yufeng. Research on high productivity I/O architecture and imple-menting technology for multi-core microprocessor chip[D]. Changsha: National University of Defense Technology, 2010.
[7] 赵改善, 李剑峰, 王于静, 等. 网格计算技术及其在石油勘探开发中 的应用前景[J]. 石油物探, 2005, 44(5): 413-420. Zhao Gaishan, Li Jianfeng, Wang Yujing, et al. Grid computing and prospect in petroleum E&P[J]. Geophysical Prospecting for Petrole, 2005, 44(5): 413-420.
[8] Wang P H, Collins J D, Chinya G N, et al. Architecture and program-ming environment for a heterogeneous multi-core multithreaded system[C]// Proceedings of the 2007 ACM SIGPLAN Conference on Program-ming Language Design and Implementation, PLDI''07, 2007.
[9] Nickolls J, Buck I. NVIDIA CUDA software and GPU parallel comput-ing architecture[C]// Proceeding of the 6th International Symposium on Memory Management, 2007.
[10] 赵改善. 我们需要多大和多快的计算机[J]. 勘探地球物理进展, 2004, 27(1): 22-28. Zhao Gaishan. How big and fast computers can meet our needs[J]. Progress in Exploration Geophysics, 2004, 27(1): 22-28.
[11] Thakkar S, Huff T. Internet Streaming SIMD Extensions[J]. Computer, 1999, 32(12): 26-34.
[12] 迟学斌, 赵毅. 高性能计算技术及其应用[J]. 中国科学院院刊, 2007, 22(4): 306-313. Chi Xuebin, Zhao Yi. High performance computing technology and its applications[J]. Bulletin of Chinese Academy of Sciences, 2007, 22(4): 306-313.
[13] 赵改善. 地球物理高性能计算的新选择：GPU计算技术[J]. 勘探地球 物理进展, 2007, 30(5): 399-405. Zhao Gaishan. New alternative to geophysical high performance com-puting: GPU Computing[J]. Progress in Exploration Geophysics, 2007, 30(5): 399-405.
[14] Lee S, Min S J, Eigenmann R. OpenMP to GPGPU: A compiler frame-work for automatic translation and optimization[C]// Proceedings of the 14th ACM SIGPLAN symposium on Principles and practice of parallel programming, 2009.
[15] Lam M S, Rothberg E E, Wolf M E. The cache performance and opti-mizations of blocked algorithms[C]// Proceedings of 4th International Conference on Architectural Support for Programming Languages and Operating Systems, 1991.
[16] Lee S, Eigenmann R. OpenMPC: Extended OpenMP Programming and Tuning for GPUs[C]// Proceedings of the 2010 ACM/IEEE International Conference forHigh Performance Computing, Networking, Storage and Analysis, 2010.
[17] 李博, 刘红伟, 刘国峰, 等. 地震叠前逆时偏移算法的CPU/GPU实施 对策[J]. 地球物理学报, 2010, 53(12): 2938-2943. Li Bo, Liu Hongwei, Liu Guofeng, et al. Computational strategy of seis-mic pre-stack reverse time migration on CPU/GPU[J]. Chinese Journal of Geophysics, 2010, 53(12): 2938-2943.
[18] 张军华, 臧胜涛, 单联瑜, 等. 高性能计算的发展现状及趋势[J]. 石 油地球物理勘探, 2010, 45(6): 918-925. Zhang Junhua, Zang Shengtao, Shan Lianyu, et al. Development status and trends for high performance computing[J]. Oil Geophysical Pros-pecting, 2010, 45(6): 918-925.
[19] 赵毅, 朱鹏, 迟学斌, 等. 浅析高性能计算应用的需求与发展[J]. 计 算机研究与发展, 2007, 44(10): 1640-1646. Zhao Yi, Zhu Peng, Chi Xuebin, et al. A brief view on requirements and development of high performance computing application[J]. Jour-nal of Computer Research and Development, 2007, 44(10): 1640-1646.
[20] 赵改善. 地球物理软件技术发展趋势与战略研究[J]. 勘探地球物理 进展, 2010, 33(2): 77-86. Zhao Gaishan. Developing trend of geophysical software technology and strategy research[J]. Progress in Exploration Geophysics, 2010, 33 (2): 77-86.
