A DEA Based Hybrid Algorithm for Bi-objective Task Scheduling in Cloud Computing

Task scheduling in cloud computing has attracted enormous attentions for its wide use in academic and industrial domains, and plays an important role in improving resource utilization and meeting QoS requirements of users. However, task scheduling is a representative NP-hard problem. Therefore, many heuristic and meta-heuristic methods have been presented to solve this problem considering many factors, such as turnaround time, execution cost, energy consuming. In this paper, we propose a meta-heuristic based algorithm HDEA to optimize turnaround time and monetary cost for task scheduling in cloud computing. This algorithm is based on a prevalent meta-heuristic, Differential evolution algorithm (DEA) and several optimization policies. In comparison with standard DEA, HDEA uses two methods to generate initial population, adopts a new mutation strategy, an adaptive parameter adjustment strategy and several local search methods with the purpose of getting better solutions. Experiments show that compared with two representative evolutionary algorithms, HDEA generates better solutions and shows competitive performance.