The Applied Problems Solving Technology Based on Distributed Computational Subject Domain Model: a Decentralized Approach
Авторы: Bychkov I.V., Oparin G.A., Bogdanova V.G., Pashinin A.A.
Журнал: Материалы XII Междунар. конф. "Параллельные вычисл. технологии" (ПаВТ'2018, Ростов-на-Дону, 2–6 апреля 2018г.)
Отчётный год: 2018
Издательство: Издат. центр ЮУрГУ
Местоположение издательства: Челябинск
Аннотация: The complexity of modern heterogeneous distributed co mputing systems prevents the widespread use of their capabilities by subject area experts and contributes to the development of specialized approaches to the creation of distributed applications that relieve the end user from the need to study the details of technical implementation. Despite the variety of such tools, some issues remain relevant. These issues include the integration of a knowledge-based approach to application development and decentralized management of application's services execution, wit ch more preferable in distributed computational networks for reliability and scalability. Thus, over the past two decades, there has been an increasing interest in developing software tools for organizing distributed computing in application packages based on decentralized management. We offer the new technology of solving the applied problems on the distributed computational model of the subject domain based on the integration of a service-oriented approach, multiagent systems, and self-organization mecha nisms in agent systems, in which all control actions are formed by a group of intellectual agents. The subject domain model is distributed the way that each agent has limited knowledge of both the capabilities of other system agents and the topology of the network. The communication model of the multi - agent system assumes only paired interactions between agents at each stage of the problem solving . This technology is demonstrated on the example of the distributed computational system creating for solving qu alitative analysis problems of autonomous binary dynamic systems having practical application in gene regulatory networks.
Индексируется WOS: 0
Индексируется Scopus: 0
Индексируется РИНЦ: 1
Публикация в печати: 0
Добавил в систему: