RUS  ENG
Full version
JOURNALS // Informatsionnye Tekhnologii i Vychslitel'nye Sistemy // Archive

Informatsionnye Tekhnologii i Vychslitel'nye Sistemy, 2018 Issue 2, Pages 44–59 (Mi itvs302)

INFORMATION PROCESSING METHODS

The software for the ground displacements processing based on massively parallel processing system Apache Spark

S. E. Popov, V. P. Potapov

The Institute of Computational Technologies of SB RAS (ICT SB RAS), Novosibirsk, Russia

Abstract: The Institute of Computational Technologies of SB RAS (ICT SB RAS), Novosibirsk, Russia The article devoted to the developing the software package for the processing radar images. It consider the ability of the visualization, configuration and running algorithms of main stages of the Persistent Scatterer method. The integration with the massively parallel processing system had shown the fast execution of calculations of the ground displacement algorithm. The paper includes main scheme of data streams routing in the Apache Spark tasks to demonstrate the network data swapping between system components in the real-time calculations. The software implementation presented as a web portal based on ReactJS+Redux components, including the automatic downloading and updating the Sentinel-1A radar database within native RESTful API. Using the approach of the Apache Spark code development paradigm allowed achieving the high performance in low execution time of calculation stages.

Keywords: monitoring of the ground displacements, radar interferometry, massively parallel processing, radar satellite imagery.

DOI: 10.14357/20718632180204



Bibliographic databases:


© Steklov Math. Inst. of RAS, 2024