Introduction
5G commercialized in 2019 using the 5G Non-Standalone (NSA) architecture, which connects 5G base stations with the existing Evolved Packet Core (EPC) to offer eMBB-oriented services. In the 5G Standalone (SA) architecture, the 5G network will include the new 5G core to enable full support of 5G use cases that not only increase throughput for mobile broadband services, but also offer Ultra-Reliable Low-Latency Communications (URLCC) that may also connect to edge computing platforms to support innovative services such as augmented or virtual reality (AR/VR,) streaming games, autonomous driving, and smart factories.
As mobile traffic volumes exploded on LTE, the mobile core began migrating to the Network Functions Virtualization (NFV) architecture, in which network functions are virtualized and provided as software entities running on commercial off-the-shelf (COTS) servers instead of dedicated network appliances. Furthermore, as the 5G mobile core evolves to enable a great variety of services, the 5G core is further leveraging the Cloud concept by migrating to a cloud-native core, in which network functions are modularized and containerized to enable highly flexible scaling and function lifecycle management.
Evolving to Cloud-Native (Source: Samsung)
The cloud-native core provides capabilities that allow the network to adapt to changing demands and support new services with minimal interactions required by operational teams.
What is Cloud-Native?
Cloud-native can be described as a combination of best practices that have been seen from companies such as Netflix, Twitter, Alibaba, Uber, Facebook, and alike. Practices include, but are not limited to, continuous deployment, containers, and microservices to help achieve the elastic scaling capabilities, speed of introducing new functionality, and increased automation needed to cater for an unpredictable competitive landscape. So, the overall goal is to be able to adapt and adapt quickly and cost-efficiently.
Also, we can define Cloud-native is an approach to building and running applications that fully exploit the benefits of the cloud computing model. For communications service providers (CSPs), applications are network functions. Examples of these functions include the Session Management Function (SMF) and the User Plane Function (UPF).
Cloud-Native
The following is a high-level description of some important cloud-native principles:
- Microservices are services that work together as a distributed system. Microservices are small, focused, and autonomous.
- Containers are a method of virtualization that bundles an application with all its dependencies—required executables, binaries, libraries and configuration files, for example—into a singular package.
- Continuous integration/continuous delivery (CI/CD) is a DevOps technique that supports frequent code changes or service updates (sometimes daily) while verifying that those changes do not negatively affect service functionality.
- Dynamic cloud-based management takes advantage of automation, containers as a service (CaaS), and other orchestration tools to keep the network up and running 24/7.
Why Cloud-Native?
As the operator’s business paradigm shifts to service-oriented businesses, the challenge faced by network operators preparing for the 5G era is to build an environment in which they can launch services promptly and upgrade the services frequently, as influenced by market demands. Telco operators began introducing the Network Functions Virtualization (NFV) environment to reduce CAPEX and boost service agility.
With the transformation to NFV allowing network functions to run as software on COTS servers, operators have been able to manage network capacity and optimize CAPEX to accommodate traffic/subscriber growth. On the other hand, the complexity of network operations has increased due to vendor-specific solutions that manage their implementations of Virtualized Network Functions (VNFs). In the 5G era, where everything from small sensors to high-speed automobiles is connected, innovation is a key factor in the network transformation.
For MNOs, going cloud-native delivers several benefits:
- Reduced capital and operational expenditure for better cost-efficiency.
- Faster time-to-market for new services.
- Scalability to grow services according to demand.
- Flexibility to update services and develop new ones quickly.
- Resiliency so that if one instance of a service fails, another one can quickly take its place.
5G Core Cloud Native Network
5G Standalone (5G SA) New Radio and Core (5G NR, 5GC) require unprecedented levels of automation across an end-to-end network to fulfill the needs of new services and applications. The 5G Core Network must be flexible, programmable, and distributed in nature so that it provides the necessary agility to shorten time-to-market and provides the utmost performance and efficiency gains. No longer a matter of selling the same service to everyone, the 5G Standalone Core differentiates the service provider, because 5G’s value is in varied custom services delivered from the Core.
Cloud-Native 5G Core (Source: Ericsson)
The core network in 5G follows a Service-Based Architecture (SBA) where network elements advertise and provide services that can be consumed by other elements in the core via APIs. This allows for the adoption of web-scale technologies and software into telecom networks.
Web-scale technologies rely primarily on open source software and bring in significant automation.
SBA is one of the cornerstones of the new 5G core: it applies IT network technologies to mobile services, enabling greater service innovation and offering sizeable benefits to operators. However, Service Based Architecture is a major departure from previous generations of mobile networks.
Service-Based Architecture (SBA)
The SBA offers a host of benefits, including:
- Deploys as containers orchestrated by Kubernetes, allowing the core to run on non-proprietary infrastructure.
- Lets new software vendors plug-and-play their NFs for the best breeds approach.
- Enables network slicing, with dynamic and efficient resource utilization.
- Simplifies operations using an application programming interface (APIs).
- Leverages the use of harmonized protocols such as HTTP/2 and its well-developed security mechanisms.
- Facilitates seamless integration of third-party applications with the core network.
Which are the main functions in 5G SBA?
5G Core has about two dozen network functions: Access and Mobility Management Function (AMF), Application Function (AF), Authentication Server Function (AUSF), Binding Support Function (BSF), CHarging Function (CHF), Network Data Analytics Function (NWDAF), Network Exposure Function (NEF), Network Repository Function (NRF), Network Slice Selection Function (NSSF), Network Slice Specific Authentication and Authorization Function (NSSAAF), Policy Control Function (PCF), Session Management Function (SMF), UE radio Capability Management Function (UCMF) and Unstructured Data Storage Function (UDSF), Unified Data Repository (UDR), User Plane Function (UPF) and Unified Data Management (UDM).
Conclusion
5G is not just about new devices, new use cases, and higher speeds. The core network needs to be modernized to be able to support demanding performance requirements. eMBB,mMTC, and URLLC are all different use cases that can’t be satisfied by a monolithic architecture.
5G Core needs to be flexible, agile, and scalable. The user plane and control plane needs to be scaled independently. Traffic handling must be optimized. Network operators must be able to quickly launch new services. This calls for virtualization, a software-driven approach, and adopting web protocols and cloud technologies. The network must be composed of loosely coupled network functions that can be managed independently but interfaced efficiently via mature and scalable technologies.
