Exploring design choices for Next Generation 5G/6G Core Design improves the chances of enhancing the flexibility and longevity of the core investment (i.e., CAPEX, Design, etc). 
Achieving efficient integration of functions across various entities and cutting down on system state and processing is essential. To meet the requirements, it is important to eliminate any duplicate functions. This is crucial to achieving the desired outcome, ensuring the task is accomplished without unnecessary repetition. 
Compared to 2G and 3G networks, 4G and 5G have reduced the number of nodes involved in the user plane. 4G has three nodes (eNB, SGW, and PGW), while 5G has only two (gNB and UPF). This increased flexibility is made possible by allowing independent scaling and placement of different functions and facilitating the creation of new services. 
Functional placement can be optimized using various criteria, such as latency, security, resilience, and energy efficiency.
Three significant developments have made it possible to have flexible systems:
The separation of the user plane and control plane
The virtualization of the Core
The cloud-native implementation of the Core 
With the evolution of 5G RAN to vRAN, a similar shift occurs with separating the base station control and user plane functions, cloud-native implementation, and centralized placement. 
In the future, the service-based architecture approach of the 5G Core will extend to the RAN. This will bring the Core user plane functions closer to the edge due to increasing traffic volume and lower latency requirements, allowing for network simplification. The vision for 6G includes a ‘Lower Layer Function’ entity with all latency-critical air interface-related RAN functions, ‘User Plane Micro Services’ (i.e., UPMS), and ‘Control Plane Micro Service’ (i.e., CPMS) functional entities with all higher-layer RAN and Core capabilities as micro-services. These micro-services may be located in the het-cloud in a disaggregated fashion, possibly with a local and central instance serving different use cases. 
This RAN-CORE convergence and functional optimization will enable highly specialized RAN and slice-specific RAN and more advanced and affordable introduction of new services and devices with different radio capabilities and dedicated software stacks, decreasing time to market and network total cost of ownership.
