AN IMPROVED MODEL OF OPTIMAL USE OF RESOURCES OF A CLUSTER SYSTEM OF MILITARY ASSIGNMENT BASED ON NASH EQUILIBRIUM
DOI:
https://doi.org/10.17721/2519-481X/2023/79-15Keywords:
finding the optimum, cluster, game theory, Nash equilibrium, fuzzy logicAbstract
The article solves the actual scientific task of optimizing the use of server resources of a military cluster system in the process of processing a large number of client requests. The setting of this scientific task is due to the need to eliminate the dependence of the efficiency indicators of the existing methods of balancing the load of cluster systems on the quality of training, as well as a rather narrow approach of the analyzed scientific studies to the study of the influence of the properties of verifiability, robustness and the guarantee of optimality on the final decision-making result. A study of existing approaches to the optimal use of cluster system resources was conducted. Based on the results of the analysis, methods based on the game-theoretic scientific-methodical apparatus were selected as the most appropriate approach to the formal description of the functional interaction of servers with client requests, as well as an approach that does not require the generalization of empirical experience (learning).
A model of the optimal use of resources of the military cluster system is proposed, which is based on the hybridization of the application of the game-theoretic approach and the theory of fuzzy logic. The essence of the proposed model is to determine the necessity of decomposition of the received task into subtasks, initiated by the client request, by means of the theory of fuzzy logic with the aim of further finding the optimal distribution of the use of cluster system servers based on the Nash equilibrium, taking into account the generalizing indicator of the current level of their workload. The obtained results are considered as a basis for the implementation of new adaptive systems for balancing the load of nodes of the cluster system, the application of which allows optimal use of the resources of the cluster system of military purpose in a stable balance.
References
1. Fesokha, V.V., Neroznak, Ye.I., Sova, O.Ya. and Nesterov, O.M. (2023), “Metod adaptyvnoho balansuvannia navantazhennia v klasternykh systemakh viiskovoho pryznachennia na osnovi rivnovahy Nesha” [The method of adaptive load balancing in cluster sustems of military purposes based on Nash Equilibrium], Collection of the scientific papers of the Centre for Military and Strategic Studies of the National Defence University of Ukraine named after Ivan Cherniakhovskyi, No. 3(76), pp. 101-110. https://doi.org/10.33099/2304-2745/2022-3-76/101-110
2. Begam, G.S., Sangeetha, M. and Shanker, N.R. (2022), “Load Balancing in DCN Servers through SDN Machine Learning Algorithm”, Arabian Journal for Science and Engineering, Vol. 47, pp. 1423-1434.
3. Klots, Y.P., Stefanovytch, K.Y., Shakhoval, Y.S. and Demeshko, V.I. (2019), “Dynamichne balansuvannia trafiku mizh dekilkoma provaideramy” [Dynamic traffic balance between several providers], Herald of Khmelnytskyi national university, No. 4(275), pp. 62-67. https://www.doi.org/10.31891/2307-5732-2019-275-4-62-67
4. Naz N.S., Abbas S., Khan M.A., Abid B., Tariq N. and Khan M.F. (2019), “Efficient Load Balancing in Cloud Computing using Multi-Layered Mamdani Fuzzy Inference Expert System”, International Journal of Advanced Computer Science and Applications, Vol. 10 No. 3, pp. 569-577.
5. Yang, Y., Li, W., Gulliver, T.A. and Li, S. (2020), “Bayesian Deep learning-Based Probabilitic Load Forecasting in Smart Grids”, IEEE Transactions on Industrial Informatics, Vol. 16 No. 7, pp. 4703-4713.
6. Breiman L., Friedman J., Olsen R. and Stone C. (1984), Classification and Regression Trees, illustrated, reprint, Wadsworth International Group, CA.
7. Campesato O. (2020), Python 3 for Machine Learning, Mercury Learning and Information, Dulles, VA.
8. Li Yu., Wang S., Hong X. and Li Yo. (2018), “Multi-objective Task Scheduling Optimization in Cloud Computing based on Genetic Algorithm and Differential Evolution Algorithm”, 37th Chinese Control Conference (CCC), Wuhan, China, July 25-27, 2018, pp. 4489-4494.
9. Chen L., Wu K. and Li Y. (2014), “A Load Balancing Algorithm Based on Maximum Entropy Methods in Homogeneous Clusters”, Entropy, Vol. 16 No. 11, pp. 5677-5697.
10. Load Balancer For Always-On Application Experience - Kemp (2023), “Load Balancing Algorithms, Types and Techniques - Kemp”, available at : https://kemptechnologies.com/load-balancer/load-balancing-algorithms-techniques (accessed 7 June 2023).
11. Apache Hadoop (2022), “MapReduce Tutorial”, available at : https://hadoop.apache.org/docs/r1.2.1/mapred_tutorial.html (accessed 7 June 2023).
12. Rotshtein, O.P. (1999), “Intelektualni tekhnolohii identyfikatsii: nechitki mnozhyny, henetychni alhorytmy, neironni merezhi : monohrafia” [Intelligent identification technologies: fuzzy sets, genetic algorithms, neural networks], UNIVERSUM-Vinnytsia, Vinnytsia, 320 p.
13. Mitiushkin, Yu.I., Mokin, B.I. and Rotshtein, O.P. (2002), “Soft Computing: identyfikatsiia zakonomirnostei nechitkymy bazamy znan: monohrafiia” [Soft Computing: identification of regularities by fuzzy knowledge bases], UNIVERSUM-Vinnytsia, Vinnytsia, 145 p.
