Nowadays, cloud computing is a technology that eludes the cost of delivery while providing pay-per-use scalability and elasticity to accessible resources. Workflow scheduling is the main challenge in Infrastructure-as – a-Service (IaaS) clouds to meet the increasing demand of computing power for large-scale scientific workflow applications. A Hybrid Multi-Objective Particle Swarm Optimization For Scientific Workflow Scheduling As workflow scheduling is part of the NP-complete problem, meta-heuristic approaches are more preferred option. Users often specified deadlines and budget constraints for scheduling these workflow applications over cloud resources.

Conclusion

Over the years, many researchers have focused their attention with a single objective on the cloud workflow scheduling problem. The decision-maker’s goal, however, is multiple and prefers Pareto’s set of optimal solutions when considering real-life applications. To solve the cloud workflow scheduling problem, we proposed the multi-objective Hybrid Particle Swarm Optimization (HPSO) algorithm based on non-dominance sorting procedure. It is a combination of multi-objective Particle Swarm Optimization algorithm and heuristic list-based. Its performance is analyzed using three conflicting goals of makepan, total cost and energy consumption under deadlines and budget constraints.