| 中期数值预报系统T213L31在IBM/SP高性能计算机上的建立 |
| |
| 引用本文： | 陈起英 金之雁 伍湘君 姚明明 AliMechentel.中期数值预报系统T213L31在IBM/SP高性能计算机上的建立[J].应用气象学报,2004,15(5):523-533. |
| |
| 作者姓名： | 陈起英 金之雁 伍湘君 姚明明 AliMechentel |
| |
| 作者单位： | 1.国家气象中心, 北京 100081 |
| |
| 摘 要： | 在引进欧洲中期天气预报中心 (ECMWF) 的全球谱模式的基础上，通过对原模式的分析改造，首次以分布与共享相结合的方式在国家气象中心IBM/SP高性能计算机上实现了全球谱模式的高效运行。采用调整向量长度、优化程序设计、完善消息传递机制和实现MPI与OpenMP的混合并行编程等方法，减少模式的通信量、计算量和内存的使用量，提高了计算效率。实现了在T213L31分辨率条件下，10天预报可以在3 h之内完成，达到业务对时限的要求。建立了与T213L31全球谱模式相配套的最优插值（OI）并行处理分析系统，解决了由于观测站点在全球不均匀分布所带来的计算负载不均衡问题。在此基础上，实现了T213L31全球资料同化与预报系统并建立了相应的自动作业监控系统。
|
| 关 键 词： | 数值天气预报 全球谱模式 分布共享并行 全球分析同化 |
| 收稿时间： | 2003-05-26 |
| 修稿时间： | 2003年5月26日 |
Foundation of Medium-range Numerical Forecast System T213L31 on High Performance Computer IBM/SP |
| |
| Chen Qiying,Jin Zhiyan,Wu Xiangjun,Yao Mingming,Ali Mechentel.Foundation of Medium-range Numerical Forecast System T213L31 on High Performance Computer IBM/SP[J].Quarterly Journal of Applied Meteorology,2004,15(5):523-533. |
| |
| Authors: | Chen Qiying Jin Zhiyan Wu Xiangjun Yao Mingming Ali Mechentel |
| |
| Affiliation: | 1.National Meteorological Center, Beijing 1000812.Chinese Academy of Meteorological Sciences, Beijing 1000813.IBM TJ Watson Research Center, New York 10598, USA |
| |
| Abstract: | Using combined method of distributed and shared memory parallelization, new global spectral model based on ECMWF is run on high performance computer IBM/SP with great efficiency in National Meteorological Center (NMC) for the first time. The model performance is improved greatly by means of changing vector length, optimizing program design, ameliorating message passing mechanism, realizing joint application of distributed and shared memory parallelization, to decrease communication, calculation and memory consumption. Now 10 day forecasts of the model with resolution T213L31 can be finished within 3 hours so as to satisfy operational time requirement. At the same time, the parallel Optimization Interpolation (OI) analysis system matching T213L31 global spectral model is constructed, and the problem of calculation load imbalance produced by spatial heterogeneousness of global observation stations is resolved, so that T213L31 global data assimilation and forecast system is built based on the above work, and so is corresponding automatic job control and watching system. |
| |
| Keywords: | Numerical weather forecasts Global spectral model Distributed and shared memory parallelization Global analysis and assimilation |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
| 点击此处可从《应用气象学报》浏览原始摘要信息 |
|
点击此处可从《应用气象学报》下载全文 |
