A dynamic service allocation algorithm in mobile edge computing

As an emerging technology, Mobile Edge Computing (MEC) is introduced to reduce network delay and provide context-aware services. MEC servers are located in close proximity to users, enabling users to seamlessly access services running on edge facilities. However, capacity and bandwidth constraints of MEC servers limit the number of services to be deployed. Therefore, an important problem is how to allocate services properly within capacity and bandwidth constraints, which is known as an NP-hard problem. In this paper, we focus on the services allocation problem in MEC and try to find trade-offs between average network delay and load balance. With temporal locality information, we allocate services by Pareto-based optimal k-Medoids and generate approximate optimal service allocation policies. In our simulation environment, compared with some traditional heuristic algorithms, our approach can reduce the variance of the load on MEC servers by 18.9% with nearly same network delay.