Multiprocessor tasks scheduling. Fuzzy logic approach / Dariusz Dorota // W: New advances in dependability of networks and systems : proceedings of the Seventeenth International Conference on Dependability ... więcej

Multiprocessor tasks scheduling. Fuzzy logic approach / Dariusz Dorota // W: New advances in dependability of networks and systems : proceedings of the Seventeenth International Conference on Dependability of Computer Systems : DepCoS-RELCOMEX, June 27-July 1, 2022, Wrocław, Poland [Dokument elektroniczny] / eds. Wojciech Zamojski [et al.]. – Cham : Springer, 2022. – (Lecture Notes in Networks and Systems, ISSN 2367-3389 ; 484). – S. 50-62. – Tryb dostępu: https://link.springer.com/chapter/10.1007/978-3-031-06746-4_6. – Bibliogr. 15 poz. – Abstr. – doi: 10.1007/978-3-031-06746-4_6. – ISBN 978-3-031-06746-4 (online). – ISBN 978-3-031-06745-7 (print)

The article concerns a problem of optimizing the execution of so-called multiprocessor tasks under the conditions of problem data uncertainty. From the implementation point of view, this is the case, where ... więcej

The article concerns a problem of optimizing the execution of so-called multiprocessor tasks under the conditions of problem data uncertainty. From the implementation point of view, this is the case, where tasks require simultaneous access to many independent processors in order to realize an embedded system with increased dependability through redundancy. The uncertainty of data may concern parameters such as, for example, task execution times, task submission deadlines, number of tasks, etc., which have been perceived in this paper as a fuzzy data. Appropriate fuzzy approach has been described and analysed. There was also provided a short overview of approaches dedicated for modeling and scheduling problem solving with particular emphasis on uncertain data, especially a fuzzy approach. Starting from the deterministic off-line algorithm for multiprocessor tasks, a variant of the algorithm using fuzzy logic approach was formulated. The approach presented in this article is innovative compared to previous work focusing on uncertainty in the scheduling of multi-processor tasks. Then, the results of the experiments were presented in tables and in Gantt charts.