DOI: 10.17587/prin.16.115-121

Formation of Plans for Organizing Parallel Computing on Processors with Very Long Instruction Word

V. M. Bakanov, Professor, e881e@mail.ru, Higher School of Economics, National Research University, Moscow, 101000, Russian Federation

Corresponding author: Valery M. Bakanov, Professor, Higher School of Economics, National Research University, Moscow, 101000, Russian Federation, E-mail: e881e@mail.ru

The article examines the issues of developing rational methods for developing schedules (execution plans) of programs for processors with a very long machine word (VLIW-architecture). The implementation of the task is carried out on the basis of an approach based on the analysis of the information structure of algorithms and targeted equivalent transformations of the tiered-parallel form of algorithm graphs. The developed software system allows implementing various (heuristic) methods for constructing a parallel program execution schedule and quantitatively evaluating and comparing the computational complexity of these methods. Quantitative results of applying some methods for constructing rational plans for parallel execution of classical data processing algorithms on processors with an extra-long machine word are presented.

Keywords: analysis of the information structure of a program, graph representations of an algorithm, rational parameters for executing parallel programs, methods for constructing a rational plan for executing a parallel program

pp. 115—121

For citation:
Bakanov V. M. Formation of Plans for Organizing Parallel Computing on Processors with Very Long Instruction Word, Programmnaya Ingeneria, 2025, vol. 16, no. 3, pp. 115—121. DOI: 10.17587/prin.16.115-121 (in Russian).

References:

1. Voevodin Vl. V. Computational Mathematics and the Structure of Algorithms (textbook, 10 lectures), Moscow, Izdatel'stvo MGU, 2006, 112 p. (in Russian).
2. AlgoWiki. Otkrytaya enciklopediya svojstv algoritmov / Eds V. Voevodin, Dzh. Dongarra, available at: http://algowiki-project.org (date of access 30.11.2024) (in Russian).
3. Bakanov V. M. Practical analysis of algorithms and efficiency of parallel computing, Moscow, Probel-2000, 2023, 198 p. (in Russian).
4. Codrescu L. Architecture of the Hexagon 680 DSP for mobile imaging and computer vision, 2015 IEEE Hot Chips 27 Symposium (HCS), Cupertino, CA, USA, 2015, pp. 1—26. DOI: 10.1109/HOTCHIPS.2015.7477329.
5. Prihozhy А. А. Analysis, transformation and optimization for high performance parallel computing, Minsk, BNTU, 2019, 229 p.
6. Gehry M., Johnson D. Computing machines and difficult tasks, World, Book on Demand, 2012, 420 p.
7. Ierusalimschy R. Programming in Lua, Third Edition, PUC-Rio, Brasil, Rio de Janeiro, 2013, 348 p.