Izvestiya Vysshikh Uchebnykh Zavedenii. Elektronika home

Publication of the journal

The section is currently being updated

Load balancing and resource allocation management in data centers represent critical challenges for distributed systems processing substantial volumes of data traffic. In this work, a mathematical model and an algorithmic framework for load management optimization in distributed computing systems are proposed. The advantages and disadvantages of cache-aware and cache-unaware algorithms were identified. To address the issue of unbalanced load distribution across proxy servers and computing nodes, a novel mathematical model for load balancing has been developed, specifically optimized for large-scale data transmission through bandwidth-constrained proxy infrastructures. The model incorporates dynamic analysis of critical parameters: the attribute structure of incoming messages, current queue depth of the mth server, and its maximum throughput capacity. Based on this model, an adaptive multithreaded load-balancing algorithm is proposed that simultaneously considers three efficiency factors: temporal processing costs, predicted message handling duration (weighted by task priority), and resource expenditure characteristics for data transmission. This approach not only ensures equitable load distribution but also minimizes latency for high-priority messages, which is a crucial capability in bandwidth-constrained network environments.

Key words: load management, distributed computing system, data center, cache-unaware algorithms, cache-aware algorithms, proxy server
Published in: INFORMATION-COMMUNICATION TECHNOLOGIES
Bibliography link: Aung Kyaw Myo, Portnov E. M., Volkov A. S., Si Thu Thant Sin, Kokin V. V. Method for load management in data processing centers in distributed computing systems. Izv. vuzov. Elektronika = Proc. Univ. Electronics. 2026;31(2):212–224. (In Russ.). https://doi.org/10.24151/1561-5405-2026-31-2-212-224.
Financial source: the work has been supported by the Russian Science Foundation (project no. 24-29-00530).

Aung Kyaw Myo
National Research University of Electronic Technology, Russia, 124498, Moscow, Zelenograd, Shokin sq., 1

Evgeny M. Portnov
National Research University of Electronic Technology, Russia, 124498, Moscow, Zelenograd, Shokin sq., 1

Alexander S. Volkov
National Research University of Electronic Technology, Russia, 124498, Moscow, Zelenograd, Shokin sq., 1

Si Thu Thant Sin
National Research University of Electronic Technology, Russia, 124498, Moscow, Zelenograd, Shokin sq., 1

Vitaly V. Kokin
National Research University of Electronic Technology, Russia, 124498, Moscow, Zelenograd, Shokin sq., 1

1. Никишин К. И. Балансировка нагрузки данных в распределенной сети через прокси-сервер NGINX. Изв. Юго-Западного гос. ун-та. 2022;26(3):98–111. https://doi.org/10.21869/2223-1560-2022-26-3-99-112. EDN: XFZZTX.

Nikishin K. I. Load balancer of data in a distributed network via NGINX proxy server. Izv. Yugo-Zapadnogo gos. un-ta = Proc. of the Southwest State Univ. 2022;26(3):98–111. https://doi.org/10.21869/2223-1560-2022-26-3-99-112

2. Richardson B., Istiono W. Comparison analysis of Round Robin algorithm with highest response ratio next algorithm for job scheduling problems. International Journal of Open Information Technologies. 2022;10(2):21–26.

3. Сорокин В. А., Бакаев А. А. Организация передачи данных по протоколу Modbus RTU. In: Энергетика: состояние, проблемы, перспективы: материалы XII Всерос. науч.-техн. конф. (Оренбург, 28–30 окт. 2021). Оренбург: Оренбургский гос. ун-т; 2021, с. 136–139. EDN: MSMVIZ.

Sorokin V. A., Bakaev A. A. Organization of data transmission via Modbus RTU protocol. In: Energetika: sostoyanie, problemy, perspektivy: proc. of the 12th Russia-wide sci.-tech. conf. (Orenburg, Oct. 28–30, 2021). Orenburg: Orenburg State Univ.; 2021, pp. 136–139. (In Russ.).

4. Волков А. С., Баин А. М., Портнов Е. М., Семенов М. Ю., Царапкин С. Ф. Разработка метода консистентного хеширования с весовыми коэффициентами для балансировки загрузки в системе управления энергообеспечением. Перспективы науки. 2024;(3):54–58. EDN: ILOWSO.

Volkov A. S., Bain A. M., Portnov E. M., Semenov M. Yu., Tsarapkin S. F. Development of a consistent hashing method with weighting coefficients for load balancing in an energy management system. Perspektivy nauki = Science Prospects. 2024;(3):54–58. (In Russ.).

5. Kusuma P. D. Weighted Round Robin based production scheduling model for seasonal product. Journal of Theoretical and Applied Information Technology. 2020;98(11):1854–1866.

6. Пискарев А. А., Портнов Е. М. Анализ способов балансировки нагрузки приложений на NodeJS. Аспирант и соискатель. 2021;(4):44–48. EDN: XVVDOZ.

Piskarev A. A., Portnov E. M. Analysis of ways of load balancing NodeJS applications. Aspirant i soiskatel’. 2021;(4):44–48. (In Russ.).

7. Аунг Чжо Мьо, Маршалов В. Н., Портнов Е. М. Разработка методики прогнозирования нагрузки в распределенной вычислительной системе. Перспективы науки. 2020;(11):187–192. EDN: UZSAKL.

Aung Kyaw Myo, Marshalov V. N., Portnov E. M. Development of load forecasting methods in a distributed computing system. Perspektivy nauki = Science Prospects. 2020;(11):187–192. (In Russ.).

Information about the publication

Downloads: 0

UDK: 004.75
Publication type: Scientific article
Source lang: Russian
DOI: 10.24151/1561-5405-2026-31-2-212-224
RISC id: UHXUXX

Download: File not found (1.31 Mb) JATS XML