Dynamic Task Scheduling Methods in Heterogeneous Systems: A Survey

systems create unlimited opportunities and challenges in the fields of parallel processing, design of algorithms, and partitioning and mapping of parallel tasks. In mapping of parallel tasks, scheduling plays an important role. The scheduling decisions are based on the dynamic parameters that may change during run time. The goal of scheduling is to utilize all the processors with minimum execution time by proper allocation of tasks to the processors. Task scheduling achieves high performance in the heterogeneous systems. A Parallel application can be represented by a Directed Acyclic Graph (DAG), which represents the dependency among tasks, based on their execution time and communication time. Previous studies reveal that many researchers have designed efficient scheduling algorithms with different parameters on multiprocessor systems. This paper investigates different aspects in scheduling and issues in various levels of the heterogeneous systems.