The rapid development of Internet of Things applications, along with the limitations of cloud computing due mainly to the far distance between Internet of Thing devices and cloud-based platform, has promoted a newly distributed computing platform based on collaboration between cloud computing and fog computing. Fog computing helps to reduce transmission latency and monetary cost for cloud resources, while cloud computing helps to fulfill the increasing demands of large-scale compute-intensive offloading applications. In this article, we study the tradeoff issue between the makespan and cloud cost when scheduling large-scale applications in such a platform. We propose a scheduling algorithm called Cost-Makespan aware Scheduling heuristic whose major objective is to achieve the balance between the performance of application execution and the mandatory cost for the use of cloud resources. Additionally, an efficient task reassignment strategy based on the critical path of the directed acyclic graph modeling the applications is also proposed to refine the output schedules of the Cost-Makespan aware Scheduling algorithm to satisfy the user-defined deadline constraints or quality of service of the system. We also verify our proposal by extensive simulations, and the experimental results show that our scheduling approach is more cost-effective and achieves better performance compared to others.
Today, the rapid growth of the large-scale applications conduces to a difficult challenge to individuals as well as the commercial organizations due to the limitation of the local computing resources. Meanwhile, the extension of the local computing platforms requires a huge investment about both finance and human power. Therefore, the use of nearlyunlimited resources and on-demand scaling of Cloud computing is considered as a potential solution to address the problems above. However, since Cloud computing is built from a payas-you-go model, the novel application scheduling approaches are challenged to guarantee the high efficiency of application execution and the reasonable cost for renting Cloud resources. In this paper, we present a novel framework built from the combination between the computing resources on Cloud computing and the computing components in the local systems. The key component of this framework is the Cost with Finish Timebased scheduling algorithm, which provides the balance between performance of application schedule and the mandatory cost for the use of Cloud resources. The experiments and comparison with the other scheduling approaches demonstrated the potential benefits of our proposed algorithm.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.