M.V. Chernenkiy - Associate Professor, Department - System of Information Processing and Management?, Bauman Moscow State Technical University E-mail: chernen @ bmstu.ru I.V. Tikhonov - Post-graduate Student, Department - System of Information Processing and Management?, Bauman Moscow State Technical University E-mail: ilya.tikxonov@optimalmngmnt.com

The purpose of this article is to define indicators to assess the feasibility of using BigData platform for simulation modeling. The features of the construction platforms and BigData development of these platforms in time. Advantages and bottle-necks in the implementation of a variety of tasks on these platforms. The article notes that the system of simulation and simulation models of large systems are on the threshold of a new stage of evolution. It is noted that a key factor for the transition to the creation and use of simulation models on the BigData platform is not large amounts of data used in the simulation, and the number of concurrent processes. The remainder of this article discusses some of the principles of the simulation process. For the simulation model takes a set of processes that describe the functioning of a system and implemented on a separate server each., next, we introduce the concept map of process states of the system, allowing to define concepts such as server load, the total load of servers, the maximum load of all servers, the coefficient of efficiency of use of servers, the effective number of servers. After that we determined the composition and structure of simulation processes. The presentation is based on the theory describing the processes, which is based on the algorithmic model of the process (AMP). For example, consider the concept class of concurrent processes and defined the concept of quasiparallel process. Using the map of the states of system processes we can show simulation under conditions of strong interdependence of operators between all processes of the simulation model on a single server is not very different in efficiency from a multi-server implementation. However, the example mapping of process states of the system, the significant effect of using multiple servers in the simulation of a large number of loosely connected units. Such systems can be classified as models of urban road transport models, civil transport within the region, models of social networks, models of cellular networks, models of computer equipment for corporations, etc. We can say that if the conditions referred to in article, the implementation of the simulation models Big Data platform-based can give significant computational effect.

 

1. Sukhobokov A.A., Lakhvich D.S. Vlijanie instrumentarija BigData na razvitie nauchnykh disciplin, svjazannykh s modelirovaniem // EHlektronnyjj zhurnal «Nauka i obrazovanie». MGTU im. N.EH. Baumana. 2015. № 3. S. 207-240. URL: http://technomag.bmstu.ru/doc/761354.html (data obrashhenija 06.02.2016).
2. Uajjt T. Hadoop. Podrobnoe rukovodstvo. SPb.: Piter. 2013. 672 s.: il.
3. Karau H. Fast Data Processing With Spark. Packt Publishing. 2013. 120 p.
4. Exploring How HP Delivers Data Protection for Modern Infrastructures // HP, September 2015 / Buffington J., Keane M., 10 p. URL: http://h20195.www2.hp.com/v2/ GetDocument.aspx-docname=4AA6-1937ENW&doctype=analyst report& doclang=EN_US&searchquery=&cc=us&lc=en,en-us (data obrashhenija 06.02.2016).
5. xREF: System x Reference // Lenovo, 14 January 2016 / Watts D., Krutov I., 196 p. URL:https://lenovopress.com/redpxref.pdf (data obrashhenija 06.02.2016).
6. Makarov V.L., Bakhtizin A.R., Vasenin V.A., Roganov V.A., Trifonov I.A. Sredstva superkompjuternykh sistem dlja raboty s agent-orientirovannymi modeljami // Programmnaja inzhenerija. 2011. № 3. S. 2-14. URL: http://iipo.tu-bryansk.ru/ fileadmin/user_upload/_content_admins/Journal_PrI/ZHurnal_Programmnaja_Inzhenerija_2011_No_3.pdf (data obrashhenija 06.02.2016).
7. CHernenkijj V.M. Algoritmicheskaja model opisanija diskretnogo processa funkcionirovanija sistemy // EHlektronnyjj zhurnal «Nauka i obrazovanie». MGTU im. N.EH. Baumana. 2011. № 12. 77-30569/292620
8. CHernenkijj V.M. Formirovanie imitacionnogo processa na osnove algoritmicheskogo opisanija funkcionirovanija informacionnojj sistemy // EHlektronnyjj zhurnal «Nauka i obrazovanie». MGTU im. N.EH. Baumana. 2011. № 11. 77-30569/291975.