Resources distribution model of critical IT infrastructure with clear parameters based on the genetic algorithm
The detailed analysis of researches of methods and algorithms of allocation of resources of virtualized IT-infrastructures is carried out. The classic model of cloud services, which consists of three layers, is considered. It is shown that the specificity of tasks performed in critical IT infrastructures puts the developer with increased requirements for reliability, security and availability. It is determined that it is expedient to use the service IaaS for implementation of the created model. The main providers of this cloud service were analyzed, their advantages and disadvantages were determined, the best candidate for implementation was selected. The following is a detailed description of the mathematical model of resource allocation of a critical IT infrastructure with clear parameters and its use in conjunction with the genetic algorithm. The following article describes the virtual machine management model for server virtualization. The example shows how it is used to solve the problem and how it can be optimized and accelerated. Subsequently, the article details the genetic algorithm, the principle of constructing a fitness function and its main operations to solve the problem. The proposed genetic algorithm is more similar to traditional genetic algorithms. At the beginning of the algorithm, an initial population of decision-individuals is created randomly. Next, each iteration of the algorithm calculates the value of the fitness function of each individual, for each individual in the population a couple is selected to generate individuals of the next population. After that, a mutation operation is applied. In addition, the search for the best individual of the new population is searched and compared with the best individual of the previous population. Finally, for the constructed model, a number of refinements are given that allow us to use this model for a critical IT infrastructure, taking into account high availability requirements such as fault tolerance (the ability of the system to continue working after the failure of one of its elements), continuous availability (the ability of the system to continuous maintenance, regardless of the time of failure of the system's nodes) and high availability (the ability of the system to further work after the failure of one of the nodes, with possible breaks in the work). The last part of the article presents experimental researches of the proposed model of distribution of resources of critical IT infrastructure with clear parameters based on the genetic algorithm.
Full Text:PDF (Українська)
L. S. Hloba, M. A. Skulish, ta O.M. Diadenko. Mathematical Foundations of Construction of Information and Telecommunication Systems. Kyiv, Ukraine: Norita-plius, 2007.
M Gorin, “Corporate Data Center: Outside of Technology”, Connect! The world of communication, no 8, pp. 19-20, 2007.
I. Kirillov, “Commercial data centers in Ukraine: a new stage of development”, Networks and business, no. 3 (52), 2010. [Online]. Available: http://www.sib.com.ua/arhiv2010/2010_3/ statia_3_1_2010/statia_3_1_2010.htm.
L. Badger, T. Grance, R. Patt-Corner, and J. Voas, “Cloud Computing Synopsis and Recommendations. Recommendations of the National Institute of Standards and Technology”, Special Publication 800-146. NIST, 2012.
N. Bibershtein, and S. Bouz, Compass in the world of service-oriented architecture (SOA). Moskow, Russia: KUDITS-Press, 2007.
D. Krafzik, K. Banke, and D. Slama, Enterprise SOA: Service-Oriented Architecture Best Practices. Upper Saddle River, USA: Prentice Hall Professional, 2004.
R. Kaur, and A. Kaur, “A Review Paper on Evolution of Cloud Computing, its Approaches and Comparison with Grid Computing”, International Journal of Computer Science and Information Technologies, vol. 5, no. 5, pp. 6060-6063, 2014.
A. Gupta, O. Sarood, and L. Kale, “Optimizing VM Placement for HPC in the Cloud”, International Letters of Social and Humanistic Sciences, vol. 16, pp. 1-6, 2014. doi: 10.1145/2378975.2378977.
J. Joseph, and C. Fellenstein, Grid Computing. Upper Saddle River, USA: Prentice Hall Professional, 2004.
J. Nabrzyski, J. Schopf, and J. Węglarz, Grid Resource Management: State of the Art and Future Trends. Berlin, Germany: Springer, 2004.
B. Goldworm, and A. Skamarock, Blade servers and virtualization: transforming enterprise computing while cutting costs. Hoboken, USA: Wiley Publishing, 2007.
N. Ruest, and D. Ruest, Virtualization, A Beginner’s Guide. New York, USA: McGraw Hill Professional, 2009.
S. F. Telenyk, O. I. Rolik, M. M. Bukasov, A. Yu. Labunskyi,, “Models of virtual machine management with server virtualization”, Bulletin of NTUU “KPI”. Informatics, Management and Computer Science, no. 51, pp. 147-152, 2009.
D. Kholland, “Genetic Algorithms”, In the world of science, no. 9-10, pp. 32-40, 1992.
J. H. Holland, Adaptation in natural and artificial systems. An introductory analysis with application to biology, control, and artificial intelligence. London, UK: Bradford book edition, 1992.
D. E. Goldberg, Genetic algorithms in search, optimization, and machine learning. Boston, USA: Addison-Wesley, 1989.
R. Buyya, J. Broberg, and A.M. Goscinski, Cloud Computing: Principles and Paradigms. New York, USA: John Wiley & Sons, 2010.
R. Sookhtsaraei, M. Madani, and A. Kavian, “A multi objective virtual machine placement method for reduce operational costs in cloud computing by genetic”, International Journal of Computer Networks and Communications Security, no. 8, c. 250-259, 2012.
G. Wu, M. Tang, Y. Tian, and W. Li, “Energy-Efficient Virtual Machine Placement in Data Centers by Genetic Algorithm”, in Proc. of International Conference on Neural Information Processing, Lake Tahoe, 2012, pp. 315-323.
ISSN 2411-1031 (Print), ISSN 2518-1033 (Online)