对新一代卫星重力数据处理中的计算密集型任务进行分析,总结出需要采用并行计算技术提高效率的几个关键任务。对不同的并行手段进行比较,采用OpenMP并行方法,并通过算例验证并行设计方法的有效性。结果表明,并行计算能显著提高卫星重力任务的数据处理效率。
The computation-intensive tasks in the data processing for satellite gravity are analyzed and some key tasks which need to be parallelized for the performance improvement are summarized.By the comparison between the parallelization methods,the OpenMP is adopted and the validity of the parallelization is demonstrated by some examples.The results show that the parallel computing can improve the efficiency of the data processing in satellite gravity mission remarkably.