Certain investigations on particle Swarm optimization techniques for Multiprocessor scheduling problems

Abstract

The importance of scheduling has increased in recent years due to NEWLINE the growing consumer demand for variety of reduced product life cycles NEWLINE changing markets with global competition and rapid development of new NEWLINE processes and technologies Two cases of the task scheduling exist in NEWLINE literature namely independent and dependent task scheduling The NEWLINE independent or dependent tasks can be static or dynamic Scheduling static NEWLINE tasks require a priori knowledge of its execution time on the processing NEWLINE nodes Dynamic task scheduling scheme does not require a prior knowledge NEWLINE of the tasks to be scheduled The tasks considered for scheduling can be NEWLINE preemptive or non preemptive in nature The minimization of the make span NEWLINE is the intent of the task scheduling problem The thesis work comprises a snapshot NEWLINE of particle swarming techniques including variations in the algorithm and its applications in the NEWLINE area of multiprocessor scheduling Algorithm inspired from Particle Swarm NEWLINE Optimization PSO method and its variants are successfully implemented for NEWLINE multiprocessor scheduling optimization problems Its efficiency for solving NEWLINE the problem is to minimize the make span criterion Genetic Algorithm PSO NEWLINE and its variants are used to solve the task scheduling problem Static task NEWLINE scheduling is implemented for both dependent and independent tasks NEWLINE NEWLINE