METHOD FOR SOFTWARE OPTIMIZATION OF BIDIRECTIONAL DATA TRANSFER BETWEEN INFORMATION SYSTEM COMPONENTS 

Bychkov Andrey Stanislavovich, candidate of technical sciences, associate professor, sub-department of computer engineering, Penza State University (40 Krasnaya street, Penza, Russia), e-mail: a_s_bychkov@mail.ru
Kalashnikov Vitaliy Aleksandorovich, student, Penza State University (40 Krasnaya street, Penza, Russia), E-mail: mymailfordev@gmail.com
Boriskin Vyacheslav Vladimirovich, student, Penza State University (40 Krasnaya street, Penza, Russia), E-mail: s.boriskin287@gmail.com
Karamysheva Nadezhda Sergeevna, candidate of technical sciences, associate professor, sub-department of computer engineering, Penza State University (40 Krasnaya street, Penza, Russia), E-mail: karamyshevans@yandex.ru
Yurova Olga Viktorovna, candidate of technical sciences, associate professor, sub-department of computer engineering, Penza State University (40 Krasnaya street, Penza, Russia), E-mail: olyura55@rambler.ru 

004.02 

10.21685/2227-8486-2020-2-8 

Subject and goals. This article describes a method for software optimization of bidirectional data transfer between components of an information system without using vertical and horizontal scaling. The method described in the article was applied to an information system consisting of a central server and a set of client subsystems. A constant increase in the load on the central server (caused by an increase in the number of participants in the exchange of information and an increase in the volume of transmitted data) is a characteristic feature of the described information system. The method allows to distribute the load on the central server by automatic distributing the start time of the data exchange procedure between as many participants as it is needed depending on the time specified request density; thereby this eliminates the problem of peak loads on the central server that occurs as a result of mass requests from client subsystems. The application of this method on the target information system made it possible to avoid urgent expensive upgrades of the architecture of the central part whose would have been necessary due to the growth of the number of new participants in the information exchange and the expansion of the information array. This method also allows to deploy the target information system in the new environment on relatively small computing capacities and delay the scaling procedure. 

software load distributing, data transfer optimization, information systems, finite state machine