DOI: 10.17587/prin.9.318-331

The Web-Service for the Classification of Seismic Events Based on Apache Spark API

S. E. Popov, e-mail: popov@ict.sbras.ru, R. Ju. Zamaraev, e-mail: zamaraev@ict.sbras.ru, I. E. Kharlampenkov, e-mail: kharlampenkov@ict.sbras.ru, Institute of Computational Technologies SB RAS, Novosibirsk, 630090, Russian Federation

Corresponding author: Popov Semion E., Senior Researcher, Institute of Computational Technologies SB RAS, Novosibirsk, 630090, Russian Federation. E-mail: popov@ict.sbras.ru

The article describes the key points of the service development process for fast automatic classification of seismic signals based on diagnostic templates. The software solutions for the preliminary signal processing and algorithm of parallel computations of the mathematical model for the development of final conclusions on the basis of rating voting are presented. Their integration with the Apache Spark distributed computing system is shown. Performance tests of the classification algorithm for a set of daily signals in various software environments were conducted. It is shown that the launch of the classification algorithm in the context of massively-parallel execution of the problem gives a gain in productivity (a decrease in the operating time) by several tens of times. The service was developed using the libraries React and Redux. The NodeJS platform is used as the runtime environment.

Keywords: web service, distributed computing, Apache Spark, classification of seismic events

pp. 318–331

For citation:
Popov S. E., Zamaraev R. Ju., Kharlampenkov I. E. The Web-Service for the Classification of Seismic Events Based on Apache Spark API

The study was carried out with the financial support of the Russian Foundation for Basic Research within the framework of a research project No. 18-07-00013A