Proof-of-Concept of Distributed Optimization of Micro-Services on Edge Computing for Beyond 5G

3G and 4G mobile services have become widespread in our lifecycle, and expectations for a better quality of throughput/latency experience are being placed to 5G era. In recent two years, 5G pilot services have been launched to deal with the drastic increase in mobile devices and application types. However, since 5G adopts a higher frequency than conventional 3G/4G, base stations must be densified. Therefore, virtualization resources for softwarization in various technologies (e.g., Virtualized RAN (Radio Access Network), Virtualized Core, Virtualized Applications) are more appreciated. In this research, we focus on edge computing, which is one of the cores of Beyond 5G, to utilize the virtualization resources. Critical potentials of edge computing are traffic reduction, low latency, and high security for mobile networks. However, various studies were made based only on numerical analysis or parts of the trial of the experiment due to a lack of technical preparation for end-to-end (E2E) environments (e.g., devices, RAN, Core, applications, human resources) with software/hardware flexibility. To show the state-of-the-art edge computing for Beyond 5G (B5G), we established a true E2E B5G Edge Cloud in Ookayama Campus of Tokyo Institute of Technology, Japan. In addition, focusing on the effective use of computation resources in Edge Clouds as an advantage of edge computing, we proposed a cloud-native with network function deployment architecture for computation resource sharing in different Edge Clouds. Specifically, we focused on Virtualized CU (Central Unit) in RAN and proposed to operate CU-CP (Control Plane) and CU-UP (User Plane) in different Edge Cloud. Its feasibility is validated through a measurement.