期刊文献+
共找到2篇文章
< 1 >
每页显示 20 50 100
PMODTRAN:a parallel implementation based on MODTRAN for massive remote sensing data processing
1
作者 Fang Huang Ji Zhou +3 位作者 Jian Tao Xicheng Tan Shunlin Liang Jie Cheng 《International Journal of Digital Earth》 SCIE EI CSCD 2016年第9期819-834,共16页
MODerate resolution atmospheric TRANsmission(MODTRAN)is a commercial remote sensing(RS)software package that has been widely used to simulate radiative transfer of electromagnetic radiation through the Earth’s atmosp... MODerate resolution atmospheric TRANsmission(MODTRAN)is a commercial remote sensing(RS)software package that has been widely used to simulate radiative transfer of electromagnetic radiation through the Earth’s atmosphere and the radiation observed by a remote sensor.However,when very large RS datasets must be processed in simulation applications at a global scale,it is extremely time-consuming to operate MODTRAN on a modern workstation.Under this circumstance,the use of parallel cluster computing to speed up the process becomes vital to this time-consuming task.This paper presents PMODTRAN,an implementation of a parallel task-scheduling algorithm based on MODTRAN.PMODTRAN was able to reduce the processing time of the test cases used here from over 4.4 months on a workstation to less than a week on a local computer cluster.In addition,PMODTRAN can distribute tasks with different levels of granularity and has some extra features,such as dynamic load balancing and parameter checking. 展开更多
关键词 Parallel computing message passing interface MODTRAN thermal infrared remote sensing land-surface temperature retrieval
原文传递
基于OpenCL的泛Kriging算法并行化研究
2
作者 范光松 黄方 +2 位作者 卜栓栓 JIANTao 董蕾 《测绘科学》 CSCD 北大核心 2017年第5期17-24,共8页
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... 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。 展开更多
关键词 OPENCL 泛Kriging算法 异构计算
原文传递
上一页 1 下一页 到第
使用帮助 返回顶部