A Review of Multiprocessor Directed Acyclic Graph (DAG) Scheduling Algorithms

Scheduling parallel tasks on multiprocessor systems is to allocate a set of tasks to processors such that the optimal usage of processors and accepted computation time for scheduling algorithm are obtained. In multiprocessor systems, an efficient scheduling of a parallel program onto the processors that minimizes the entire execution time is vital for achieving a high performance. The need for a scheduling algorithm arises from the requirement for most modern systems to perform multitasking. A multiprocessor system can be homogeneous (i.e. having all processors of same capability) or heterogeneous (i.e. having processors of different capabilities) nature. In literature several algorithms are proposed to find optimal solution of multiprocessor task scheduling algorithms. So, in present study we have reviewed various multiprocessor scheduling approaches to find their drawbacks so that an efficient heuristic based algorithm can be developed for obtaining optimal solution for these problems.