Title :
Coevolutionary multi-agent systems: the application to mapping and scheduling problems
Author_Institution :
Inst. of Comput. Sci., Polish Acad. of Sci., Warsaw, Poland
Abstract :
The paper addresses two closely related problems, namely the mapping and scheduling tasks of a parallel program in parallel computers. To develop new efficient algorithms to solve the aforementioned problems we propose a theoretical framework based on a model of N-person games which serves as a model of interaction of agents in a multi-agent system. We implement the multi-agent system with use of evolutionary techniques and apply it to develop parallel and distributed algorithms of dynamic mapping and scheduling
Keywords :
cooperative systems; game theory; parallel algorithms; parallel programming; scheduling; software agents; N-person games; coevolutionary multiagent systems; distributed algorithms; dynamic mapping; parallel algorithms; parallel computers; parallel program; scheduling; Application software; Computer aided manufacturing; Computer science; Concurrent computing; Cost function; Job shop scheduling; Manufacturing processes; Multiagent systems; Processor scheduling; Scheduling algorithm;
Conference_Titel :
Industrial Technology, 1996. (ICIT '96), Proceedings of The IEEE International Conference on
Conference_Location :
Shanghai
Print_ISBN :
0-7803-3104-4
DOI :
10.1109/ICIT.1996.601624
