三维变分资料同化系统并行算法设计与实现  

IMPLEMENTATION OF PARALLEL ALGORITHM OF THREE DIMENSIONAL VARIATIONAL DATA ASSIMILATION

作　　者：王玉柱[1,2] 姜金荣[1] 蔡长青[1,2] 迟学斌[1] 岳天祥[3] 

机构地区：[1]中国科学院计算机网络信息中心超级计算中心,北京100190 [2]中国科学院大学,北京100049 [3]中国科学院地理科学与资源研究所,北京100101

出　　处：《数值计算与计算机应用》2013年第3期231-240,共10页Journal on Numerical Methods and Computer Applications

基　　金：国家高技术发展计划(2010AA012301;2012AA01A309)资助项目

摘　　要：三维变分资料同化作为现在主流数值天气预报的同化方法,能够明显改善预报数据的质量.随着科学研究的逐渐深入以及科学探测仪器和计算机技术的不断发展,受计算量和内存需求量的限制,传统串行三维变分资料同化系统已无法满足高分辨率、高精确度数值预报的要求.所以,三维变分资料同化系统的并行设计与实现显得尤其重要.本文设计了混合二维区域剖分并行化方法及其通信算法库,并将其应用于国家气象局三维变分同化系统3DVAR.数值试验表明,系统128核的并行效率相对于2核高达72%,具有良好的加速效果;同时,内存需求也随处理器个数的增加而成倍减少,满足了高分辨率预报的要求.As one of the most important data assimilation methods in numerical weather pred- ication, 3DVAR （three-dimensional variational data assimilation） can improve the quality of data predication significantly. With gradual development of scientific research and the improvement of detecting instruments and computer technology, the traditional sequential 3DVAR system can no longer meet the demand for the high resolution and high accuracy nu- merical forecasting due to computational capacity and memory limit. So the parallel design and implementation of 3DVAR is very essential. In the paper, a mixed domain decomposi- tion parallel method and message communication library are applied in the 3DVAR system of the state meteorological agency. Numerical experiments show that the 3DVAR system could run with high speedup ratio and parallel efficiency.

关 键 词：三维变分资料同化 区域剖分 递归滤波 并行算法 

分 类 号：P456.7[天文地球—大气科学及气象学]