A Flower Pollination Algorithm Based Task Scheduling in Cloud Computing

Abstract

Flower pollination algorithm (FPA) is a nature inspired fascinating meta-heuristic technique, which is applicable to many real life optimization problems. Mapping of tasks on the virtual machines in cloud computing environment is a well known NP-complete problem. This paper propose a novel FPA-based algorithm to schedule the tasks on the virtual machines to minimize makespan and maximize cloud resource utilization. The proposed scheme uses an efficient pollen representation scheme and a novel multi-objective fitness function. We simulate the proposed scheme on one synthetic and two benchmark datasets of diverse configuration. The performance of the proposed scheme is compared with three other meta-heuristic based approaches, namely particle swarm optimization (PSO), genetic algorithm (GA) and gravitational search algorithm (GSA). The superiority of the proposed algorithm over other algorithms is exhibited by the simulation results.