This is a task scheduler simulator. some well-known algorithms are implemented.
Task scheduling in heterogeneous systems is an important topic and an NP-Complete problem for the high-performance execution of programs. Programs are made with a sequence of tasks that may have prerequisites. For showing the program we use DAG (directed acyclic graphs). In each graph, nodes are tasks, and edges show requirements. Each task has its own process time on each processor. Also if a task wants to run on a processor except for the processor that its parent ran, the transport time should be considered. In this paper, we purpose a new hybrid algorithm by combining a list-based algorithm with a metaheuristic algorithm. We use a list scheduling algorithm for creating an entry population. Then we try to improve schedulings by using the Simulated Annealing algorithm. The algorithm focuses on improving the length of scheduling and finding minimum makespan. The purposed algorithm is evaluated on different randomly generated DAGs, and results Compared with the well-known existing algorithms. The results of experiments show about 2% improvement in makespan.