Труды ХVII международной конференции «Аналитика и управление данными в областях с интенсивным использованием данных» (“Data Analytics and Management in Data Intensive Domains” (DAMDID))
The need for transmission and storage of large amounts of scientific data in the project space radio telescope "Radioastron" required us to organize a reliable communication channel between the tracking station in Pushchino and treatment data centre in Moscow. Managing such a communication network requires a complex approach and encompasses establishing secure access for a range of tasks. These tasks include controlling network devices, monitoring, timely replacement of equipment, software and firmware updates, data backups, as well as network infrastructure documentation.
In 2012 the multi-beam feed array became operational on the BSA FIAN radio telescope. As of July 2014 it is capable of 24-hour observation using 96 beams in declination of -8 to +42 degrees in the 109-111.5 MHz frequency band. The number of frequency bands range from 6 to 32, while the time constants range from 0.1 to 0.0125 sec. While receiving in 32 band mode (plus one common band) with a time constant of 12.5 ms (80 times per second), 33x96x80 four-byte data samples are produced per second, which equates to a daily data production of 87.5 gigabytes (up to 32 terabytes yearly). This data is an enormous opportunity for both short and long-term monitoring of various classes of radio sources (including radio transients), space weather and the Earth's ionosphere monitoring, search for different classes of radio sources, etc. The base goals of our work are: a) To obtain new scientific data on different classes of discrete radio sources, the construction thereof physical models and evolutions - obtained on the basis of continuous, round-the-clock monitoring of the radio sky in the 109-111.5 MHz frequency band and cross-analysis of data from third-party surveys of other frequencies. b) To launch data streaming on various types of high-performance computing systems, including creating a public system of distributed computing for thousands of users on the basis of BOINC technology. The BOINC client for astronomical data from the monitoring surveys of large parts of the entire sky almost has no analogies.