An improved rank-based genetic algorithm with limited iterations for grid scheduling

In most cases, the number of resources and tasks in grid computing environment is large. Accordingly, the complexity of task scheduling is significantly increased. This results very complex optimization problem. This paper proposes an improved rank-based roulette wheel selection genetic algorithm (IRRWSGA) for scheduling independent tasks in the grid environment. The modified algorithm speeds up convergence and shortens the search time, at the same time the heuristic initialization of initial population using MCT algorithm allow the algorithm to obtain a high quality feasible scheduling solution. The simulation results, show that IRRWSGA has better search performance than both IGA and standard genetic algorithms in terms of both time and quality of the schedule. IRRWSGA improves the reliability in the selection process and produces an enhanced solution. Fast reduction of makespan is a practical concern for grid environment. Real-world scheduling problems may utilize this algorithm for better results.