Cost-aware scheduling for ensuring software performance and reliability under heterogeneous workloads of hybrid cloud

摘要

Cloud computing is a rapidly growing paradigm in software engineering that offers different services. The hybrid cloud is the best choice for the enterprise to benefit by taking resources on lease from the public cloud only if private cloud resources are not sufficient. However, the key is how to provide better cloud services and improve software performance in the hybrid cloud for software engineers. In this paper, the efficient job scheduling method in the private cloud is proposed by considering the heterogeneity of hybrid cloud resources to guarantee the software performance and reliability. The experimental results show that the efficient job scheduling method can effectively reduce the average job response time and improve the system throughput. Moreover, the task scheduling method based on BP neural network in the hybrid cloud is proposed by considering both the cost and deadline constraints to ensure the quality of service (QoS) for software. The experimental results show that the task scheduling method can improve the QoS, maximize the resources utilization of private cloud and minimize the cost of hybrid cloud resources.