PARALLELISM IN LARGE SOFTWARE PACKAGES (WHY IS IT DIFFICULT TO CREATE EFFICIENT SOFTWARE)
https://doi.org/10.22405/2226-8383-2017-18-3-187-200
Abstract
In the article, types of parallelism used in architectures of modern computer systems are considered, and the ways of their manifestation in programs are described. Six paradigms of parallel programming are analyzed, and the relationship of paradigms to generations of highperformance computing systems is shown. Different methods of description and representation of parallelism based on various kinds of program models are considered. The reasons that determine challenges of developing efficient software packages for parallel computing systems are discussed. The connection between the material under discussion and the actively developed Internet encyclopedia of properties and features of AlgoWiki parallel algorithms is noted.
About the Author
V. V. Voevodin
Lomonosov Moscow State University
Russian Federation
Russian Federation
doctor of physical and mathematical sciences, professor, corresponding member of Russian Academy of Sciences, Deputy Director, Research Computing Center
References
1. Antonov, A.S., Voevodin, V.V. & Dongarra, J. 2015, “Algowiki: an Open encyclopedia of parallel algorithmic features Journal of Supercomputing Frontiers and Innovations, vol. 2, no. 1, pp. 4-18.
2. Antonov, A 2012, Tehnologii parallel’nogo programmirovanija MPI i OpenMP [Parallel Programming Technologies MPI and OpenMP]. Izdatelskij Dom MGU, Мoscow, 344p.
3. Sadovnichy, V., Tikhonravov, A., Voevodin, V. & Opanasenko V. 2013, "Lomonosov": Supercomputing at Moscow State University. In Contemporary High Performance Computing: From Petascale toward Exascale” (Chapman & Hall/CRC Computational Science), CRC Press, Boca Raton, USA, pp. 283-307.
4. Voevodin, V. 1992 Mathematical Foundations of Parallel Computing (Series in Computer Science), World Scientific Publishing Co.Pte. Ltd., Singapure, vol. 33, 364p.
5. Voevodin, V.V., Voevodin, Vl.V. 2004, Parallel’nye vychislenija [Parallel computing], BHV-Peterburg, St. Petersburg, 608p.
6. Voevodin V. “Parallel Algorithms: Theory, Practice and Education”, (Proceedings of the Joint Workshop on Sustained Simulation Performance, University of Stuttgart (HLRS) and Tohoku University), Springer International Publishing, 2016, pp. 3-10.
7. ATLAS installation guide (2016), Available at: http://mathatlas. sourceforge.net/atlas_install/ (accessed 5 December 2017).
8. Fidesys – software package for engineering analysis (2017), Available at: https://cae-fidesys.com/ru/download (accessed 5 December 2017).
9. Levin, V.A., Vershinin A.V., Sabitov D.I., Nikiforov I.V. & Pendjur D.A. 2010 “Using a supercomputer technologies in strength problems. Fidesys Package”, Supercomputer technologies in science, education and industry, Izdatelskij Dom MGU, Мoscow, vol.2., pp.161-166.
Review
For citations:
Voevodin V.V. PARALLELISM IN LARGE SOFTWARE PACKAGES (WHY IS IT DIFFICULT TO CREATE EFFICIENT SOFTWARE). Chebyshevskii Sbornik. 2017;18(3):187-200. (In Russ.) https://doi.org/10.22405/2226-8383-2017-18-3-187-200
Views:
464
Email this article 