Robust and Cost-effective Resource Allocation for Complex IoT Applications in Edge-Cloud Collaboration

The rapid increasing of the Internet-of-Things (IoT) applications make it convenient to sense and collect real-world information in our daily life. To ensure the performance of these IoT applications, researchers established an edge-cloud collaboration application system based on the multi-access edge computing (MEC) paradigm where the IoT data can be processed not only on the cloud but also on nearby edge servers. However, as the edge servers are resource-limited, we should be more careful in allocating the edge resource to the application, especially when it is composed by several micro-services. In this paper, we considered how edge-cloud cooperation can help running these service composition based IoT applications and proposed an efficient resource allocation approach to balance performance, robustness, and cost-effectiveness of IoT applications in MEC environments. We mathematically modeled the cost-effective performance optimization problem in robust edge-cloud application systems and proved the convexity of the approximated problem so that they can be solved in tractable ways with existing solvers to generate the resource allocation strategies. Meanwhile, we carried out a series of experiments to evaluate our approach. The experiment results showed that our approach was powerful in managing the performance, cost and robustness compared with representative baselines.