Methods and models of resource allocation service in load balancing clusters for data centers

The object of the research is computing clusters of cloud data centers, containing many servers, data storage systems, an input-output system interconnected by a communication network. The goal of this research is to develop methods and models for improving the performance of a data center cluster by reducing the processing time of service requests as well as reducing equipment costs due to the efficient allocation of its resources. Therefore, it is necessary to implement optimization algorithms for placing virtual machines (VMs) on physical servers in real time based on load balancing. The proposed method of resource allocation is based on an iterative greedy algorithm and a limited search procedure. Reduction in the computation time is achieved by introducing restrictions on the permissible search depth. The paper puts forward a mathematical model of resource allocation, built using the Erlang model in the form of a multi-line m-node queuing system (QS) of the M|M|m|n type with an n-seat buffer, which makes it possible to determine the main indicators of service request quality in the form of QS parameters. The efficiency of this approach was tested on a simulation model built on the basis of the system functioning statistical analysis. Its experimental study was also carried out.

1. Gnedenko B.V., Kovalenko I.N. Introduction to queuing theory. LCI Publisher; 2007. 400 p.

2. Aliev T.I. Fundamentals of modeling of discrete systems. St. Petersburg, ITMO; 2009. 363 p.

3. Feller E., Rilling L., Morin C. A scalable and autonomic virtual machine management framework for private Clouds. Proceedings of the 12th IEEE/ACMInternational Symposium on Cluster, Cloud and Grid Computing (CCGrid). 2021:482–489.

4. Ward J.S., Barker A. Cloud cover: monitoring large-scale clouds with Varanus. Journal of Cloud Computing: Advances, Systems and Applications, 2015;4:127–135.

5. Kleinrock L.. Queueing Theory. Mashinostroenie; 1979. 432 p.

6. Mochalov V.P., Linets G.I., Bratchenko N.Y., Govorova S.V. An analytical model of a corporate software-controlled network switch. Scalable Computing. 2020;21(2):337–346.

7. Боев В.Д. Компьютерное моделирование. Пособие для практических занятий, курсового и дипломного проектирования в AnyLogic7. Санкт-Петербург; 2014. 432 с.

8. Taihoon K., Soksoo K. Analysis of Security Session Reusing in Distribution Server System. Computational Science and Its Applications. ICCSA 2006; 2006. 1045 p.

9. Holland J.H. Adaptation in Natural and Artificial Systems:An Introductory Analysis with Applications to Biology,Control, and Artificial Intelligence. The MITPress, Cambridge; 1992. 211 p.

10. Хританков А.С. Модели и алгоритмы распределения нагрузки. Алгоритмы на основе сетей СМО. Информационные технологии и вычислительные системы. 2009;(3):33-48.

11. Ivanisenko I., Kirichenko L., Radivilova T. Balancer multifractal methods considering load characteristics. International Journal «Information Content and Processing». 2015;2(4):345–368.

12. Panchenko T.V. Genetic Algorithms. Astrakhan, Astrakhanskiy Universitet; 2007. 87 p.

13. Tsoy Yu.R., Spitsyn V.G. Genetic Algorithm. Tomsk, Knowledge Representation in Information Systems; 2006. 146 p.

14. Mochalov V.P., Bratchenko N.Y., Yakovlev S.V. Analytical model of object request broker based on Corba standard. Journal of Physics: Conference Series. 2018;1015(2). DOI: 10.1088/1742-6596/1015/2/022012.

15. McNab A., Stagni F., and Luzzi C. LHCb experience with running jobs in virtual machines. Journal of Physics: Conference Series. 2016;664:1–7.

16. Ward J.S., Barker A Observing the clouds: a survey and taxonomy of cloud monitoring. Journal of Cloud Computing: Advances. Systems and Applications. 2014;3:25–33.

17. Mochalov V.P., Bratchenko N.Y., Yakovlev S.V. Analytical model of integration system for program components of distributed object applications. International Russian Automation Conference, RusAutoCon 2018. 2018;8501806. DOI: 10.1109/ RUSAUTOCON.2018.8501806.

18. Computing Center of the Institute of High Energy Physics (IHEP-CC). «VCondor – virtual computing resource pool manager based on HTCondor». 2016. Доступно по: //github.com/hep-gnu/VCondor.

19. Anne-C´ecile Orgerie. When Clouds become Green: the Green Open Cloud Architecture. International Conference on Parallel Computing (ParCo). 2009;228–237.

20. Mochalov V.P., Bratchenko N.Yu., Yakovlev S.V., Gosteva D.V. Distributed management system for infocommunication networks based on TM Forum Framework. CEUR Workshop Proceedings. 2016;2254:81–93.

21. Mochalov V., Bratchenko N., Linets G., Yakovlev S. Distributed management systems for infocommunication networks: A model based on tm forum frameworx. Computers. 2019;8(2). DOI: 10.3390/computers8020045.

22. Mochalov V.P., Bratchenko N.Y., Yakovlev S.V. Process-Oriented Management System for Infocommunication Networks and Services Based on TM Forum Frameworx. Proceedings - 2019 International Russian Automation Conference, RusAutoCon 2019. 2019;8867619. DOI: 10.1109/RUSAUTOCON.2019.8867619.

23. McNab A., Love P., MacMahon E. Managing virtual machines with Vac and Vcycle. Journal of Physycs: Conference Series. 2015;664b:115–122.

24. Beloglazov R. OpenStack Neat: A Framework for Dynamic and Energy-Efficient Consolidation of Virtual Machines in OpenStack Clouds. Concurrency and Computation: Practice and Experience (CCPE). 2015;27(5):1310–1333.

25. Kuzin L.Т. Fundamentals of cybernetic models. М.: Energia; 1979. 584 p.

26. Open Grid Forum. «Open Cloud Computing Interface». 2016. Доступно по: http://occiwg.org/.

27. Balashov N., Baranov A., Korenkov V. Optimization of over-provisioned clouds. Physics of Particles and Nuclei Letters. 2016;13(5):609–612.