DYNAMIC METHOD OF DISTRIBUTED SYSTEM LOAD BALANCING EVALUATE
DOI:
https://doi.org/10.26906/SUNZ.2021.2.074Keywords:
load balancing, distributed system, multifractal traffic, resource utilization, self-similar flow, imbalanceAbstract
Subject of study is method of estimating resources of the distributed system like a part of scientific problem related to the load balancing and efficient utilization of resources of the distributed system. The paper presents a method of estimating resources of the distributed system, such as network nodes, the processor, memory, and band-width. The proposed method allows to calculate the loading of each node separately in a distributed system and the entire system. Classes of service flows taken into account in the calculation of these resources loading. The complex value of imbalance of load server entered, which taking into ac-count the weight coefficients for processor, memory, and network bandwidth. These weight coefficients allow to select the importance of each network resource (CPU, memory and bandwidth) compared with each other. Also, this method allows to calculate the imbalance of the system servers. Using the method in load balancing allows to distribute requests by servers such way that deviation of the load servers from the average value was minimal, that allow to provide higher system performance parameters (utilization efficiency) and faster processing flows. Conclusions. The work proposed a solution to the actual scientific problem of assessing the load of nodes of a distributed system. The proposed method is based on calculating the processor load, memory load, and channel bandwidth by flows of different service classes. Also introduced a complex value of server load imbalance, taking into account weights for processor, memory and network bandwidth. Accordingly, this method allows you to calculate the imbalance of all servers in the system, the average operating time for various balancing algorithms and the efficiency of using the system resources.Downloads
References
L. Kirichenko, I. Ivanisenko, T. Radivilova, Investigation of Self-similar Properties of Additive Data Traffic, CSIT-2015 X-th International Scientific and Technical Conference «Computer science and information technologies», Lviv, UKRAINE, 14 – 17 September, 2015, pp. 169-172
O. I. Sheluchin, S. M. Smolskiy, A. V. Osin, Self-Similar Processes in Telecommunications, New York : John Wiley & Sons, 2007, pp. 320.
Игнатенко Е.И., Бессараб В.И., Дегтяренко И.В. Адаптивный алгоритм мониторинга загруженности сети кластера в системе балансировки нагрузки. // Наукові праці ДонНТУ. – Вип. 21(183). – 2011. – С. 95-102.
Chen H., Wang F., Helian N., Akanmu G. User-priority guided min-min scheduling algo-rithm for load balancing in cloud computing // National Conference on Parallel Computing Tech-nologies (PARCOMPTECH). – Bangalore, 2013. – P. 1-8.
Cardellini V. A performance study of distributed architectures for the quality of web ser-vices. // Proceedings of the 34th Conference on System Sciences. – Vol. 10. – 2001. – P.213-217.
Keshav S. An Engineering Approach to Computer Networking // Addison-Wesley, Read-ing, MA. – 1997. – P. 215-217.
Liu J., Luo X., Zhang X., Zhang F., Li B. Job Scheduling Model for Cloud Computing Based on Multi-Objective Genetic Algorithm // IJCSI International Journal of Computer Science. – V.10(1). - No 3. - 2013. – P.134-139.
Kameda H., Li L., Kim C., Zhang Y. Optimal Load Balancing in Distributed Computer Sys-tems. – London: Springer, Verlag London Limited. - 1997. – 238 р.
Kaur R., Luthra P. Load Balancing in Cloud Computing // Proc. of Int. Conf. on Recent Trends in Information, Telecommunication and Computing. – Association of Computer Electronics and Electrical Engineers. – 2014. – P. 374-381.
Sviridov, A., Kovalenko, A. and Kuchuk, H. (2018), “The pass-through capacity redevelopment method of net critical section based on improvement ON/OFF models of traffic”, Advanced Information Systems, Vol. 2, No. 2, pp. 139–144, DOI: https://doi.org/10.20998/2522-9052.2018.2.24
Kovalenko, A.A. and Kuchuk, G.А. (2018), “The current state and trends of the development of computer systems of objects of critical application”, Systems of control, navigation and communication, PNTU, Poltava, No. 1 (47), pp. 110–113, DOI : https://doi.org/10.26906/SUNZ.2018.1.110
Donets V., Kuchuk N., Shmatkov S. Development of software of e-learning information system synthesis modeling process. Сучасні інформаційні системи. 2018. Т. 2, No 2. С. 117–121. DOI: https://doi.org/10.20998/2522-9052.2018.2.20
Zykov, I.S., Kuchuk, N.H. and Shmatkov S.I. (2018), “Synthesis of architecture of the computer transaction management system e-learning”, Advanced Information Systems, Vol. 2, No. 3, pp. 60-66, DOI: https://doi.org/10.20998/2522-9052.2018.3.10
Ruban, I.V., Martovytskyi, V.O., Kovalenko, A.A. and Lukova-Chuiko, N.V. (2019), “Identification in Informative Systems on the Basis of Users' Behaviour”, Proceedings of the International Conference on Advanced Optoelectronics and Lasers, CAOL 2019-September,9019446, pp. 574-577, DOI: https://doi.org/10.1109/CAOL46282.2019.9019446
Kovalenko, А. and Kuchuk H. (2018), “Methods for synthesis of informational and technical structures of critical application object’s control system”, Advanced Information Systems, 2018, Vol.2, No.1, pp. 22–27, DOI: https://doi.org/10.20998/2522-9052.2018.1.04
Roth G. Server load balancing architectures, Part 1: Transport-level load balancing. – 2008. – Режим доступа: http://www.javaworld.com/article/2077921/architecture-scalability/server-load-balancing-architectures--part-1--transport- level-load-balancing.html.
