|
Abstract.
The aim of the article is the conceptual development of mathematical models and methods for organizing parallel computations in the mode of synchronous interaction of locally distributed processes competing for the use of a limited number of copies of a structured software resource under conditions of unlimited and limited parallelism by the number of processors of a multiprocessor system. For this mode, using the mathematical apparatus of scheduling theory, linear diagrams, and arc-weighted graphs, the problems of determining the minimum total implementation time of a set of heterogeneous parallel distributed interacting competing processes are solved.
Keywords:
multiprocessor system, process, software resource, structuring, pipelining, unlimited (limited) parallelism, linear diagram, arc-weighted graph.
DOI 10.14357/20718632250306
EDN BDFXMH
PP. 58-72.
References
1. Antonov A.S., Afanasyev I.V., Voevodin Vl.V. (2021) High-performance computing platforms: current status and development trends. Computational methods and programming, vol. 22, pp. 135–177 (in Russian). 2. Babichev S.L., Konkov K.A. (2019) Distributed systems. Moscow: Yurait (in Russian). 3. Tanenbaum A., Steen М. (2023) Distributed systems. Amazon Digital Services LLC-Kdp. 4. Tel G. (2009) Introduction to Distributed Algorithms. Cambridge University Press. 5. Toporkov V.V., Emelyanov D.М. (2018) Models, methods and algorithms for planning in grid and cloud computing. Bulletin of the Moscow Energy Institute, no 6, pp. 75–86 (in Russian). 6. Kovalenko N.S. (2011) Mathematical modeling of parallel processes. Germany: LAP LAMBERT Academic Publishing GmbH & Co. KG (in Russian). 7. Pavlov P.А. (2010) Efficiency of distributed computing in scalable systems. Scientific and technical bulletins of SPbSPU: Computer science. Telecommunications. Management, no 1, pp. 83–89 (in Russian). 8. Pavlov P.А. (2010) Scalable distributed systems of competing interacting processes and their optimality. Bulletin of the State Samara Aerospace University named after Academician S.P. Koroleva, no 1, pp. 234–243 (in Russian). 9. Pavlov P.А. (2010) Organization of homogeneous competing processes in distributed pipeline processing. Control Sciences, no 3, pp. 66–72 (in Russian). 10. Pavlov P.А. (2010) Optimal structuring of software resources during pipeline distributed processing. Software products and systems, no 3, pp. 76–82. 11. Pavlov P.А. (2011) The problem of optimizing the number of processors in distributed processing. Bulletin of the State Samara Aerospace University named after Academician S.P. Koroleva, no 6, pp. 230–240 (in Russian). 12. Pavlov P.А. (2012) The optimality of software resources structuring through the pipeline distributed processing of competitive cooperative processes. International Journal of Multimedia Technology (IJMT), vol. 2, no 1, pp. 5–10. 13. Kovalenko N.S., Ovseec M.I. (2012) Problems of optimizing the number of processors and constructing an optimal layout of distributed systems. BSU Bulletin. Series 1: Physics. Mathematics. Computer Science, no 1, pp. 119–126 (in Russian). 14. Kovalenko N.S., Ovseec M.I. (2012) Asynchronous distributed computations with a limited number of copies of a structured program resource. Cybernetics and systems analysis, vol. 48, no 1, pp. 86–98. 15. Kovalenko N.S. (2012) Optimal Grouping Algorithm of Identically Distributed Systems. Programming and Computer Software, vol. 38, no 3. pp. 143–149 (in Russian). 16. Pavlov P.А. (2014) Time of implementation of asynchronous parallel processes in macropipeline concentrated processing. Problems of informatics, no 3. pp. 37–52 (in Russian). 17. Kovalenko N.S., Ovseec M.I. (2012) Asynchronous distributed computations with a limited number of copies of a structured program resource. Cybernetics and systems analysis, no 1. pp. 105–117 (in Russian). 18. Kovalenko N.S. (2024) Synchronous mode of distributed computing with continuous execution of blocks of a limited number of copies of a software resource. Software products and systems, no 1, pp. 43–53 (in Russian). 19. Pavlov P.А. (2024) On the implementation time of distributed computing in synchronous mode with a limited number of copies of a software resource. Proceedings of the Institute for System Programming of the Russian Academy of Sciences, vol. 36, no 4, pp. 81–98 (in Russian). 20. Emelichev V.A., Kovalev M.M., Kravtsov M.K. (1981) Polyhedra. Graphs. Optimization. Moscow: Nauka (in Russian). 21. Tanaev V.S., Sotskov Yu.N., Strusevich V.А. (1989) Scheduling theory. Multistage systems. Moscow: Nauka (in Russian). 22. Lazarev А.А. (2019) Scheduling theory. Methods and algorithms. Moscow: ICS RAS (in Russian).
|
