유전 알고리듬을 이용한 멀티프로세서 시스템의 선형 스케쥴링 알고리즘

초록

The availability of parallel processing systems and computer networks offer a multiple of the processing power, which can be obtained ever from a fast single processor. The objective of multiprocessor scheduling can be stated as finding schedule for allocating tasks onto the processors and their execution time so that task dependence is satisfied and the length of the schedule, known as the parallel time, is minimized. The multiprocessor scheduling is a computationally complex problem, described in computer science terms as NP-complete. In this paper, we present a scheduling method based on Genetic Algorithm(GA) for the static scheduling of directed acyclic graphs with consideration of communication costs on homogeneous scheduling to search a large solution space efficiently and effectively to find the best possible(reasonable) solution.