Research on Grid Workflow Scheduling Based on the Discrete Multi-objective Particle Swarm Optimization Algorithm

Existing grid workflow scheduling algorithms commonly suffer by one or several of the following drawbacks: most only considerations of DAG grid workflow model, not enough considerations of multi-dimensional QoS parameters and the single objective optimal scheduling. The paper presents a grid workflow scheduling algorithm DMOPSO based on the discrete multi-objective particle swarm optimization. The proposed algorithm redefines the operation rules of particle's position and velocity, and the moving equation of particle for an effective use of crowding distance in multi-objective particle swarm optimization (MOPSO_CD) algorithm according to the characteristics of discrete variable. Compared with a grid workflow scheduling algorithm OMOPSO which is rounded to discrete for position and velocity of particles in the movement equation of particles based on the original MOPSO_CD algorithm, the experimental results show that the DMOPSO algorithm obtains more Pareto optimal solutions and better mean result of solutions.