A QoS-based scheduling algorithm for instance-intensive workflows in cloud environment

Along with the rapid development of the Internet, especially the advent of cloud computing era, instance-intensive workflow applications will play an increasingly important role in people´s daily life. Most of the existing scheduling algorithms are designed just for a single complex workflow instance. This paper takes cloud workflow as a research background, completely generalizes and describes the instance-intensive workflow scheduling optimization problems based on QoS (Quality of Service) under the current architecture of cloud workflows. Then this paper classifies workflow tasks based on user´s QoS preference, and combines with staggered sub-deadlines allocation criteria, proposes a QDA scheduling algorithm suitable for instance-intensive workflows in cloud environment. Simulation results show that, compared to other algorithms, QDA algorithm can better meets the different requirements from users, and also has a much better load balancing rate.