Parallel Optimization for Large Scale Interferometric Synthetic Aperture Radar Data Processing

Interferometric synthetic aperture radar (InSAR) has developed rapidly over the past years and is considered as an important method for surface deformation monitoring, benefiting from growing data quantities improving quality. However, handing of SAR big poses significant challenges related algorithms pipeline, particularly in large-scale processing. In addition, InSAR are highly complex, their task dependencies intricate. There a lack efficient optimization models scheduling pipeline. this paper, we design parallel time-series processing based on multi-thread technology high efficiency data. These concentrate parallelizing critical that have complexity, with focus deconstructing two computationally intensive through loop unrolling. Our shown improvement 10–20 times performance. We also tool, Simultaneous Task Automatic Runtime (STAR), which utilizes flow strategy thread pool to address problem low CPU utilization resulting multiple modules STAR provides data-driven pipeline enables concurrent execution tasks, greater flexibility keep busy further improve predetermined flow. Additionally, supercomputing-based system been constructed massive scientific providing technical support nationwide measurement, accordance framework time series Using system, processed volumes 500 TB 700 5 7 days, respectively. Finally generated maps land all China.