Cost-effective scheduling precedence constrained tasks in cloud computing

Aiming at related task scheduling in cloud computing, a cost-effective precedence constrained tasks scheduling algorithm is presented. The algorithm takes into account the monetary cost and tries to fulfill a task scheduling balancing time and cost. In order to explore more possible solutions with high quality ignored by the deterministic algorithm, multi-population genetic algorithm is adopted to acquire the expected scheduling scheme. Besides, in order to improve the performance on time consuming of the algorithm, improved task duplication with monetary cost restriction is proposed. The validity of the algorithm is verified by experiments. Compared with deterministic scheduling algorithm, our algorithm lessens the cost greatly. Meanwhile, its performance on time consuming can also be guaranteed to some extent.