吉林大学学报(理学版)

• 计算机科学 • 上一篇    下一篇

基于多核CPU的卫星云导风并行反演算法

王昌帅1,2, 何丽莉1,2, 欧阳丹彤1,2, 白洪涛3   

  1. 1. 吉林大学 计算机科学与技术学院, 长春 130012;2. 吉林大学 符号计算与知识工程教育部重点实验室, 长春 130012;3. 吉林大学 公共计算机教学与研究中心, 长春 130012
  • 收稿日期:2015-05-25 出版日期:2016-05-26 发布日期:2016-05-20
  • 通讯作者: 何丽莉 E-mail:helili@jlu.edu.cn

Satellite Cloud Drift Winds Parallel Inversion Algorithm Based on Multicore CPU

WANG Changshuai1,2, HE Lili1,2, OUYANG Dantong1,2, BAI Hongtao3   

  1. 1. College of Computer Science and Technology, Jilin University, Changchun 130012, China; 2. Key Laboratory ofSymbolic Computation and Knowledge Engineering of Ministry of Education, Jilin University, Changchun 130012, China;[JP]3. Center for Computer Fundamental Education, Jilin University, Changchun 130012, China
  • Received:2015-05-25 Online:2016-05-26 Published:2016-05-20
  • Contact: HE Lili E-mail:helili@jlu.edu.cn

摘要:

针对卫星云导风反演算法规模大、 计算密集、 耗时长, 常规云导风反演串行算法效率低的问题, 根据云导风反演中各风矢计算相对独立的特征, 提出一种基于OpenMP框架的云导风反演并行算法. 通过对多组云图数据在多核CPU上反演结果进行分析表明, 该算法显著提高了云导风反演的效率.

关键词: 相关系数法, 云导风, OpenMP, 并行, 风矢

Abstract:

In view of the satellite cloud drift winds inversion has the characteristics of large scale, intensive computing and timeconsuming, conventional cloud drift winds serial inversion algorithm is very difficult to break through the bottleneck of efficiency. According to the characteristics of relatively independence between each wind vector calculation, we put forward a parallel inversion algorithm based on the OpenMP framework. Through the analysis of the inversion results of the multi group cloud data in the multi core CPU, the results show that the proposed algorithm can significantly improve the inversion efficiency of cloud.

Key words: correlation coefficient method, cloud drift wind, OpenMP, parallel, cloud vector

中图分类号: 

  • TP311