Mantra: Edge computing places Application Servers close to the user, which reduces end-to-end latency and transport bandwidth requirements.
Brief Explanation. 3GPP supports edge computing, where a local User Plane Function (UPF) close to the user is selected so that the user traffic between the device and the Application Server travels for a relatively short distance. In the traditional approach, the Application Server is located far away from the device beyond the centralized core network. Hence, compared to such traditional approach, the edge computing approach reduces the end-to-end latency. Furthermore, since the user traffic stays relatively local and does not go deep inside the core transport network (where significant traffic aggregation occurs), the overall transport bandwidth requirements are reduced. Edge computing enables 5G to support latency-sensitive applications. While ETSI has defined a functional architecture for Multi-Access Edge Computing (MEC), 3GPP is also studying an architecture. The deployment of edge computing may involve interactions among ETSI NFV architecture, ETSI MEC architecture, and 3GPP edge computing functional architecture along with the baseline 3GPP NG-RAN and 5GC.