Improving decision-making algorithms on the correctness of the states of multiversions of fault-tolerant software systems
Автор: D. I. Kovalev, P. K. Zaitsev
Журнал: Informatics. Economics. Management - Информатика. Экономика. Управление.
Рубрика: Системный анализ, управление и обработка информации
Статья в выпуске: 2(3), 2023 года.
Бесплатный доступ
The task of improving the algorithms for making decisions about the correctness of the states of multiversions is relevant today for the development of fault-tolerant software systems. The article considers a class of decision-making algorithms that are implemented in the execution environment of multi-version program modules. This execution environment includes a decision block, which, based on the principle of voting, allows you to get the correct decision at the output, filtering out the erroneous results of the execution of software versions. Advanced algorithms allocate the output results of multiversion triggering into classes or subsets, which are then analyzed for correctness. This approach is characterized by the fact that it allows you to increase the stability of the runtime environment to interversion errors. This, in turn, helps to improve the fault tolerance of software systems used in reliability-critical applications.
Software, algorithm, reliability-critical application, fault-tolerant system, decision-making
Короткий адрес: https://sciup.org/14127463
IDR: 14127463 | DOI: 10.47813/2782-5280-2023-2-3-0201-0209