Implementation of Dynamic Level Scheduling Algorithm using Genetic Operators
Prabhjot Kaur, Amanpreet Kaur · 2013
Scheduling and mapping of precedence constrained task graphs to the processors is one of the most crucial problems in parallel computing. Due to the NP- completeness of the problem, a large portion of related work is based on heuristic approaches with the objective of finding good solutions within a reasonable amount of time. The major drawback with most of the existing heuristics is that they are evaluated with small problem sizes and their scalability is unknown. In this paper, I have implemented APN Dynamic Level Scheduling algorithm by using genetic operators for task scheduling in parallel multiprocessor system including the communication delays to reduce the completion time and to increase the throughput of the system. The parameters in this paper I have used are makespan time, processor utilization and scheduled length ratio. The graphs show better results of dynamic level scheduling with genetic operators as compared to simple dynamic level scheduling algorithm.