551.513 Algorithm for computational performance improvement and processor load balancing to simulate the general atmosphere circulation

Parkhomenko V. P. (Bauman Moscow State Technical University/Computing Centre of RAS)

ATMOSPHERIC GENERAL CIRCULATION MODEL, PARALLEL IMPLEMENTATION FEATURES, NUMERICAL EXPERIMENTS.


doi: 10.18698/2309-3684-2016-3-93109


The paper analyzes some factors affecting the parallel implementation performance of the atmospheric general circulation model designed on a cluster type multiprocessor computer. It considers several modifications of the initial parallel code of this model in order to improve both its computational efficiency and processor load balancing. The numerical scheme is modified according to the time of the atmospheric general circulation model for parallel computing of dynamics and physics blocks. The proposed procedure is used along with the procedures of paralleling the dynamics and physics blocks based on decomposition of the computational domain. It allows both optimizing the processor load balancing and increasing the paralleling efficiency. The data obtained while using the scheme for the physics block load balancing allow for complication of the physics block without increasing the total computational time. The results of numerical experiments are given.


[1] Monin A.S. Vvedenie v teoriyu klimata [Introduction to the theory of climate]. Leningrad, Gidrometeoizdat Publ., 1982, 296 p.
[2] Climate Change 2013. The Physical Science Basis. The Physical Science Basis. New York, Cambridge University Press, 2013. Available at: http://www.climatechange2013.org/images/report/WG1AR5_ALL_FINAL.pdf
[3] Moiseev N.N., Aleksandrov V.V., Tarko A.M. Chelovek i biosfera [Man and the biosphere]. Moscow, Nauka Publ., 1985, 272 p.
[4] Dimitrienko Yu.I., Koryakov M.N., Zakharov A.A. Applied Mathematical Sciences, 2015, vol. 9, no. 98, pp. 4873–4880. http://dx.doi.org/10.12988/ams.2015.55405
[5] Dimitrienko Yu.I., Koryakov M.N., Zakharov A.A. Matematicheskoe modelirovanie i chislennye metody — Mathematical Modelling and Computational Methods, 2015, no. 4, pp. 75–91.
[6] Dimitrienko Yu.I., Koryakov M.N., Zakharov A.A. Finite Difference Methods, Theory and Applications. Lecture Notes in Computer Science, 2015, vol. 9045, pp. 161–168. DOI 10.1007/978-3-319-20239-6_15
[7] Dimitrienko Yu.I., Zakharov A.A., Koryakov M.N., Syzdykov E.K. Izvestiya vysshikh uchebnykh zavedeniy. Mashinostroenie — Proceedings of Higher Educational Institutions. Machine Building, 2014, no. 3, pp. 23–34.
[8] National Aeronautics and Space Administration: Earth Systems Science: A Closer View. Report of the Earth Systems Science Committee NASA Advisor Council NASA, Washington, NASA, 1988, 208 p.
[9] Belov P.N., Borisenkov E.P., Panin B.D. Chislennye metody prognoza pogody [Numerical methods of weather forecast]. Leningrad, Gidrometeoizdat Publ., 1989, 375 p.
[10] Parkhomenko V.P., Lang T.V. Journal of Computer Science and Cybernetics, 2013, vol. 29, no. 2, pp. 138–148.
[11] Parkhomenko V.P. Realizatsiya globalnoi klimaticheskoi modeli na mnogoprocessornoi EVM klasternogo tipa. Parallelnye vychislitelnye tehnologii [The implementation of a global climate model on a cluster type multiprocessor computer. Parallel computing technologies]. Trudy Mezhdunarodnoy nauchnoy konferentsii “Parallelnye vychislitelnye tekhnologii 2009” [Proceedings of the International scientific conference “Parallel computational technologies”, 2009] (Nizhny Novgorod, 30 March – 3 April 2009). Chelyabinsk, Publishing House of the SUSU, 2009, 839 p.
[12] Parkhomenko V.P. Vestnik MGTU im. N.E. Baumana. Ser. Estestvennye nauki — Herald of the Bauman Moscow State Technical University. Series Natural Sciences, 2011, spec. iss. “Mathematical modeling”, pp. 186–200.
[13] Parkhomenko V.P. Matematicheskoe modelirovanie i chislennye metody — Mathematical Modelling and Computational Methods, 2015, no. 1, pp. 94–108.
[14] Parkhomenko V.P. Matematicheskoe modelirovanie i chislennye metody — Mathematical Modelling and Computational Methods, 2014, no. 2, pp. 115–126.
[15] Parkhomenko V.P. Inzhenernyy zhurnal: nauka i innovatsii — Engineering Journal: Science and Innovation, 2012, no. 2. DOI 10.18698/2308-6033-2012-2-45
[16] Parkhomenko V.P. Inzhenernyy zhurnal: nauka i innovatsii — Engineering Journal: Science and Innovation, 2013, no. 9. DOI 10.18698/2308-6033-2013-9-962
[17] Parkhomenko V.P. Sea Ice Cover Sensitivity Analysis in Global Climate Model. Research Activities in Atmospheric and Oceanic Modelling. Geneva, World Meteorological Organization, 2003, vol. 33, pp. 7.19–7.20.
[18] Arakawa A., Lamb V. Methods in Computational Physics, 1977, vol. 17, pp. 174–207.
[19] Parkhomenko V.P. Problemy realizatsii i funktsionirovaniya globalnoi klimaticheskoi modeli na parallelnykh vychislitelnykh sistemakh [Problems of implementation and functioning of the global climate models on parallel computer systems]. Trudy IV Mezhdunarodnoi konferentsii “Parallelnye vychisleniya i zadachi upravleniya” (Moscow, 27–29 Oct., 2008) [Proceedings of the IV International
conference Parallel computations and control problems (Moscow, 27–29 Oct., 2008)]. Moscow, V.A. Trapeznikov Institute of Control Sciences, Russian Academy of Sciences, 2008, pp. 122–141.
[20] Voevodin V.V., Voevodin Vl.V. Parallelnye vychisleniya [Parallel computations]. St. Petersburg, BHV-Petersburg Publ., 2002, 608 p.
[21] Climate Change 2007: The physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel Climate Change, Paris, 2007, 989 p.


Parkhomenko V. Algorithm for computational performance improvement and processor load balancing to simulate the general atmosphere circulation. Маthematical Modeling and Coтputational Methods, 2016, №3 (11), pp. 93-109



Download article

Колличество скачиваний: 150