范光松,黄方,卜栓栓,JIAN Tao,董蕾

• 1:电子科技大学资源与环境学院
• 2:电子科技大学大数据研究中心遥感所
• 3:Center for Computation & Technology Louisiana State University

摘要(Abstract)：

针对泛Kriging插值算法在大量数据处理时的高耗时问题,该文从异构平台主机端与OpenCL设备端的交互方式入手,采用OpenCL异构平台开发语言进行泛Kriging算法并行化实现研究,解决了其在进行大数据量处理时数据存储、数据交互、多设备调度等一系列问题。在K20Xm平台上使用不同的数据集和参数对并行泛Kriging算法进行测试,实验结果表明:与Intel Xeon E5-2670CPU平台相比,并行程序插值部分加速比达到40倍以上,整体并行程序加速比达到了18倍。

关键词(KeyWords)： OpenCL;泛Kriging算法;异构计算

Abstract:

Keywords:

基金项目(Foundation): 教育部中央高校基本科研业务费专项资金项目;; 中国科学院遥感与数字地球研究所项目;; 四川省应急测绘保障与地质灾害监测工程技术研究中心开放研究基金项目(K2014B003);; 国家博士后基金项目(2011M501400)

作者(Author): 范光松,黄方,卜栓栓,JIAN Tao,董蕾

DOI: 10.16251/j.cnki.1009-2307.2017.05.004

参考文献(References)：

• [1]BRIGGS D J,COLLONS S,ELLIOTT P,et al.Mapping urban air pollution using GIS:a regression-based approach[J].International Journal of Geographical Information Science,1997,11(7):699-718.
• [2]WU Q,XU H,ZOU X.An effective method for 3Dgeological modeling with multi-source data integration[J].Computers and Geosciences,2005,31(1):35-43.
• [3]张军.空间插值算法研究及其在遥感数据模拟中的应用[D].成都:成都理工大学,2013.(ZHANG Jun.Research of spatial interpolation algorithm and its application in hyperspectral data simulation[D].Chengdu:Chengdu University of Technology,2013.)
• [4]陈欢.气候资源插值算法在多核环境下的并行计算研究[D].长沙:湖南大学,2012.(CHEN Huan.Interpolation of climatic resources in multi-core research on parallel computing environment[D].Changsha:Hunan University,2012.)
• [5]KERRY K E,HAWICK K A.Kriging interpolation on high-performance computers[C]//roceedings of the International Conference and Exhibition on High-Performance Computing and Networking,LNCS.Springer Berlin:Heidelberg,1998:429-438.
• [6]GAJRAJ A,JOUBERT W,JONES J.A parallel implementation of kriging with a trend[EB/OL].[1997-11-01].http://www.osti.gov/energycitations/servlets/purl/544698-DhOroO/webviewable/.
• [7]PEDELTY J A,MORISERRS J T,SMITH J A,et al.High performance geostatistical modeling of biospheric resources[C]//American Geophysical Union Fall Meeting.[S.l.]:[s.n.],2004:1-4.
• [8]LU K,GODDARD S.GRASS-based high performance spatial interpolation component for spatial decision support systems[C]//Proceedings of the FOSS/GRASS Users Conference.[S.l.]:[s.n.],2004:1-13.
• [9]STRZELCZYK J,PORZYCKA S,LE′SNIAK A.Analysis of ground deformations based on parallel geostatistical computations of PSInSAR data[C]//In Proceedings of 17th International Geoinformatics Conference.[S.l.]:[s.n.],2009:1-6.
• [10]CRESSIE N,JOHANNESSON G.Fixed rank kriging for very large spatial data sets[J].Journal of the Royal Statistical Society:Series B(Statistical Methodology),2008,70(1):209-226.
• [11]陈鹏,王乘,任波.粒度可调的并行Kriging地层空间插值算法[J].小型微型计算机系统,2009,30(1):152-155.(CHEN Peng,WANG Cheng,REN Bo.Granularity-adjustable parallel kriging interpolation for stratified stratum[J].Journal of Chinese Computer Systems,2009,30(1):152-155.)
• [12]陈欢,谢健.基于OpenMP的Kriging插值算法研究[J].计算机科学,2012,39(6):392-395.(CHEN Huan,XIE Jian.Kriging interpolation algorithm based on OpenMP[J].Computer Science,2012,39(6):392-395.)
• [13]王涛.“天河二号”超级计算机[J].科学,2013,65(4):52-52.(WANG Tao.“Tianhe-2”Supercomputer[J].Science,2013,65(4):52-52.)
• [14]马俊峰.基于OpenCL的多GPU并行计算的研究与应用[D].哈尔滨:哈尔滨理工大学,2014:1-53.(MA Junfeng.Research and application of Multi-GPU parallel computing based on OpenCL[D].Harbin:Harbin University of Science and Technology,2014:1-53.)
• [15]赵加强.基于OpenCL的稀疏矩阵向量乘优化[D].长春:吉林大学,2012:1-48.(ZHAO Jiaqiang.Optimizing sparse matrix-vector multiplication based on OpenCL[D].Changchun:Jilin University,2012:1-48.)
• [16]FERREIRINHA T,NUNES R,AZEVEDO L,et al.Acceleration of stochastic seismic inversion in OpenCL-based heterogeneous platforms[J].Computers&Geosciences,2015(78):26-36.
• [17]HANZER F.Spatial interpolation of scattered geoscientific data[EB/OL].[2012-02-03].http://uni-graz.at/~haasegu/Lectures/GPU_CUDA/WS11/hanzer_report.pdf.
• [18]STONE J E,GOHARA D,SHI G.OpenCL:A parallel programming standard for heterogeneous computing systems[J].Computing in Science&Engineering,2010,12(1-3):66-73.
• [19]候景儒,潘汉军,张树泉.多元地质统计学的基本理论与方法[J].北京科技大学学报,1992,14(2):115-122.(HOU Jingru,PAN Hanjun,ZHANG Shuquan.Basic theory and methods of multivariate geostatistics[J].Journal of University of Science and Technology Beijing,1992,14(2):115-122.)
• [20]李亮亮,依艳丽,凌国鑫,等.地统计学在土壤空间变异研究中的应用[J].土壤通报,2005,36(2):265-268.(LI Liangliang,YI Yanli,LING Guoxin,et al.Utilization of geostatistics in soil spatial variability[J].Chinese Journal of Soil Science,2005,36(2):265-268.)
• [21]李文石,姚宗宝.基于阿姆达尔定律和兰特法则计算多核架构的加速比[J].电子学报,2012,40(2):230-234.(LI Wenshi,YAO Zongbao.Multicore architecture speedup computation based on Amdahl’s law and Rent’s rule[J].Acta Electronica Sinica,2012,40(2):230-234.)