摘要：The Task Scheduling in multiprocessor system is also known as multiprocessor scheduling. It is used in large number of applications from scientific and engineering to commercial problems. The major objective of task scheduling is to minimize the program's execution time. The task scheduling is represented by a directed acyclic graph (DAG). There are basically two types of task scheduling algorithms: Deterministic or Static and Non deterministic or dynamic scheduling algorithm. In this paper, we have focused only static scheduling algorithms and it is further classified into heuristic based and guided random search based. In this paper, we have taken, Bounded Number of Processors (BNP) class scheduling algorithms. It is a classification of heuristic based. The BNP scheduling algorithm consists of four scheduling algorithms: Highest Level First with Estimate Time (HLFET) algorithm, Modified Critical Path (MCP) algorithm, Earliest Time First (ETF) algorithm and Dynamic Level Scheduling (DLS) algorithm. We have studied of Bounded Number of Processors (BNP) class scheduling algorithms and find the scheduling length of each algorithm. The scheduling length is used in four matrices: Speedup, Efficiency, Load Balance and Normalized Scheduling Length (NSL). Finally, we have compared these four scheduling algorithm based on four matrices.
