Cybernetics And Systems Analysis logo
Editorial Board Announcements Abstracts Authors Archive
Cybernetics And Systems Analysis
International Theoretical Science Journal
UDC 004.4
S. Pogorilyy1, M. Slynko2


1 Taras Shevchenko National University of Kyiv, Kyiv, Ukraine

sdp@univ.net.ua

2 Taras Shevchenko National University of Kyiv, Kyiv, Ukraine

maxim.slinko@gmail.com

MODELING AND VERIFICATION METHODS FOR APPLICATION DESIGN
IN HETEROGENEOUS ARCHITECTURES

Abstract. An application design methodology for massively parallelized systems (GPGPU systems) focused on the algorithmic design phase is proposed. Two design sub-steps are considered: creation of the formal specification of the system; and its research and verification. The use of mathematical apparatuses SAA/SAA-M as well as transition systems is proposed for the first step. Advantages and specifics of using network and automaton models for the second step are given. NVIDIA CUDA architecture computation model analysis is proposed using Petri nets (network model) and using linear-temporal logic formulas and Buchi automaton (automaton model).

Keywords: transition systems, systems of algorithmic algebras, GPGPU-systems, Petri nets.



FULL TEXT

REFERENCES

  1. Pogorilyy S.D., Kryvyi S.L., Slinko M.S. Design and model substantiation of applications based on video adapters. Upravlyayushchye systemy y mashyny. 2018. N 4. P. 46–56. https://doi.org/ 10.15407/usim.2018.04.0046.

  2. Kryvyi S., Porogilyy S., Slynko M. Transition systems as method of designing applications in GPGPU technology. Problems in Programming. 2018. N 2–3. P. 12–20.

  3. Anisimov A.V., Pogorilyy S.D., Vitel D.Yu. About the issue of algorithms formalized design for parallel computer architectures. Applied and Computational Mathematics. 2013. Vol. 12, N 2. P. 140–151.

  4. Pogorilyy S.D., Maryanovsky V.A., Boyko Y.V., Vereshchinsky O.A. Investigation of parallel circuits of the Danzig algorithm for computing systems with shared memory. Mathematical machines and systems. 2009. N 4. P. 27–37.

  5. Pogorilyy S.D., Slynko M.S. Research and development of Johnson’s algorithm parallel schemes in GPGPU technology. Problems in Programming. 2016. N 2–3. P. 105–112.

  6. Pogorilyy S.D., Shkulipa I.Yu. A conception for creating a system of parametric design of parallel algorithms and their software implementations. Cybernetics and Systems Analysis. 2009. Vol. 45, N 6. P. 952–958.

  7. Dijkstra E.W., Buxton J.N., Randell B. Software engineering techniques. Report on a Conference Sponsored by the NATO Science Committee. (27–31 October 1969, Rome, Italy). Rome, Italy, 1969. P. 16.

  8. Boyko Yu.V., Kryvyi S.L., Pogorilyy S.D. et al. Methods and latest approaches to the design, management and application of high-performance IT infrastructures [in Ukrainian]. Kyiv: Kyiv University, 2016. 447 p.

  9. Albert E., Lanese I. Formal techniques for distributed objects, components, and systems. Proc. 36th IFIP WG 6.1 International Conference, FORTE 2016, held as Part of the 11th International Federated Conference on Distributed Computing Techniques, DisCoTec 2016, (June 6–9, 2016, Heraklion, Crete, Greece). Heraklion, Crete, Greece, 2016. 275 p.
© 2020 Kibernetika.org. All rights reserved.