A network slicing algorithm for cloud-edge collaboration hybrid computing in 5G and beyond networks

Abstract

Edge computing can effectively provide a promising solution with ultra-low latency for a variety of user-oriented services in 5G and beyond networks. The existing studies, however, are unable to provide a flexible resource allocation strategy for various network slices with different user demands in edge computing. This paper proposes a new network slicing algorithm for cloud–edge collaboration hybrid computing (CECHC) in 5G and beyond networks. This algorithm designs a flexible and effective resource allocation strategy primarily for three network slices based on diversified requirements. Moreover, the CECHC model is designed with optimally distributed units (DUs) and centralized units (CUs) deployments to improve storage capability and computing power. This allows for more convenient function partitioning for different network slices. To validate the performance of the proposed algorithm, a series of agent-based simulations are conducted in various network models. The experimental results demonstrate that the proposed algorithm deployed in CECHC outperforms other network models, including fog computing (FC), multi-access edge computing (MEC), and the original CECHC models. It provides the lowest latency for various network slices and achieves all successful runs with different storage capabilities and computing power.