Оцінка надійності паралельних обчислювальних систем при виконанні заданого комплексу взаємопов'язаних робіт

2019;
: сс. 15-23
Автори:
1
Національний університет «Львівська політехніка», кафедра електронних обчислювальних машин

Під час проектування паралельної обчислювальної системи для виконання заданого комплексу взаємопов'язаних робіт важливою характеристикою є її надійність. Але крім основних показників надійності обчислювальної системи, забезпечених технічними засобами, існує ще показник надійності системи, який оцінює придатність паралельної обчислювальної системи до вирішення конкретних задач за визначений час. Це оцінювання безпосередньо пов'язане з ефективністю використання обчислювальних систем.

[1] Kuznetsov P. A. On the issue of analysis of the effectiveness of systems with full redundancy. Bulletin of SibSAU, t. 16, No. 2, p. 326-330, 2015. https://doi.org/10.3103/S096709121505006X

[2] Pullum L. L. Software fault tolerance techniques and implementation. Artech House, 2001. 360 p.

[3] Roganov V. R., Grishko A. K., Kochegarov A. K. Three Approaches to evaluating the Performance of active Reservation Systems. DOI 10.21685/2307-4205-2019-2-2.

[4] Chu W. W., Leung K. K. Module replication and assignment for real-time distributed processing system // "Proc IEEE". 1987. 75. N 5. pp. 547-562. https://doi.org/10.1109/PROC.1987.13772

[5] Khritankov A. S. Mathematical model of performance characteristics of distributed computing systems. Computer science, management, economics. WORKS OF MIPT. 2010. Volume 2, No. 1 (5), p. 110-115.

[6] Ivutin A. N., Larkin E. V. Prediction of the execution time of the algorithm. Magazine. News of TSU. Technical science. Issue number 3/2013. C. 301-315.

[7] Bocharov P. L., Ignatushchenko V. V. Mathematical models and methods for evaluating the effectiveness of parallel computing systems on complexes of interrelated works // Tez. report international conf, "High-Performance Computing Systems in Management and Scientific Research", Alma-Ata, 1991.

[8] Margalitashvili A. L. Investigation of the effectiveness of the functioning of parallel computing resources on given complexes of interrelated works, Abstract of Cand. dis. M.: In-t prbblem management, 1990.

[9] Bocharov P. L., Preydunov Yu. V. Estimation of the execution time of a complex of works on a parallel computational system // System analysis and computer science. Sat scientific papers. M.: Publishing house DN, 1991. C. 29-41.

[10] Ingatushchenko V. V. Organization of structures for controlling multiprocessor computing systems. Moscow: Energoatomizdat, 1984.

[11] Kumar S., Cohen P. R. Towards a fault-tolerant multi-agent system architecture. In: Proceedings of the Fourth International Conference on Autonomous Agents. ACM, 2000, pp. 459-466. DOI:10.1145/336595.337570. https://doi.org/10.1145/336595.337570

[12] Guessoum Z., Briot J. P., Faci N. Towards Fault-Tolerant Massively Multiagent Systems. In: Massively Multi-Agent Systems I. Springer Berlin Heidelberg, 2005. P. 55-69. (Ser. Lecture Notes in Computer Science; vol. 3446). DOI: 10.1007/11512073_5. https://doi.org/10.1007/11512073_5

[13] Serugendo G. D.M., Romanovsky A. Designing Fault-Tolerant Mobile Systems. In: Scientific Engineering for Distributed Java Applications. Springer Berlin Heidelberg, 2003. P. 185-201. (Ser. Lecture Notes in Computer Science; vol. 2604). DOI: 10.1007/3-540-36520-6_17. https://doi.org/10.1007/3-540-36520-6_17

[14] Mellouli S. A Reorganization Strategy to Build Fault-Tolerant Multi-Agent Systems. In: Advances in Artificial Intelligence. Springer Berlin Heidelberg, 2007. P. 61-72. (Ser. Lecture Notes in Computer Science; vol. 4509). DOI: 10.1007/978-3-540-72665-4_6. https://doi.org/10.1007/978-3-540-72665-4_6

[15] Ignatushchenko V. V., Klushin Y. S. Prediction of the implementation of complex software systems on parallel computers: direct stochastic modeling // Automation and Remote Control. 1994. N 12, p. 142-157.

[16] Klushin, Y. S. Prediction of the implementation of complex software systems on parallel computers // Proc. Report Second Ukrainian Conference on Automatic Control "Automation-95". Lviv, 1995, vol. 2, p. 100.

[17] Ignatushchenko V. V., Klushin Yu. S. Forecasting the implementation of complex software systems on control parallel computers: exact methods // Scientific works of the International Symposium "Automated Control Systems", Tbilisi: ed. Intellect, 1996, p. 23-28.

[18] Klushin Y. S. Evaluation of the effectiveness of various dispatching disciplines for reducing the time to perform complex software systems on parallel computing systems / Bulletin of National University "Lviv Polytechnic" No. 413. Computer engineering and information technology. Lviv: NU "LP", 2000. p. 19-23.

[19] Gross, D., Miller, D., Transition Markov processes // Operations Research. 1984. Vol. 32. No. 4. P. 334-361. https://doi.org/10.1287/opre.32.2.343

[20] Reibman A. L., Trivedi K. S. Numerical transient analysis of Markov models // Computers and Operations Research. 1988. Vol. 15. No. 1. P. 19-36. https://doi.org/10.1016/0305-0548(88)90026-3

[21] Klushin, Y. S. Software implementation of mathematical models, methods and algorithms for estimating the time of execution of complex software complexes in multiprocessor computer systems. Bulletin of NU "Lviv Polytechnic" № 905. Computer systems and networks. Lviv: NU "LP", 2018. https://doi.org/10.23939/csn2018.905.073

[22] Klushin, Y. S. Improving the accuracy of estimating the execution time of folding software systems in multiprocessor computer systems for belt stochastic modeling. Bulletin of NU "Lviv Polytechnic" No. 881. Computer systems and networks. Lviv: NU "LP", 2017. Methods for real-time object searching and recognizing in video images on ios mobile platform