基于Spark的GNSS网基线向量并行化处理

Parallel Processing of GNSS Network’s Baseline Vectors Based on Spark

摘要: 针对传统单机处理大规模复杂GNSS观测数据效率低的问题,引入了Spark大数据集群,利用子网划分思想并调用GAMIT软件对GNSS网基线向量进行解算,实现了并行化计算。实验结果表明,在保证解算精度与整体解算在同一量级上的同时,提高了执行效率,性能优于整体解算,较好地满足了大规模复杂GNSS数据处理需求。

Abstract: In order to solve the problem that the traditional single machine processing largescale and complex GNSS observation data are low efficiency, the Spark large data cluster is introduced, the subnet partition idea and the GAMIT software are used to calculate the GNSS network’s baseline vector, and the parallel computing is realized. The experimental results show that the efficiency is improved and the performance is better than the whole solution,and the accuracy is in the same order, which satisfies the demand of largescale and complex GNSS data processing.