5G networks are expected to provide virtually-unlimited-gigabit and ultra-reliable connections to people and objects, when and where it matters, supporting diverse use cases with an extremely demanding range of requirements (in terms of latency, throughput, reliability, coverage and security, cost targets, etc.).
But building a network that supports a broad array of new services, all of which can be set up, scaled and torn down at a moment’s notice – that presents several challenges. Unprecedented operational agility is required to allow services to be rapidly deployed, dynamically adapted, and continuously and seamlessly assured. In addition, the network’s performance, coverage and capacity should be constantly assured to satisfy the service requirements.
5G runs on a virtualized infrastructure and is designed with a fully programmable approach to software and microservices, capable of supporting diverse use-cases. This makes the 5G network exponentially more complex to operate and manage! When we add distributed clouds, heterogenous networks, the 10X densification of RAN sites as well as the hundreds of thousands of network functions - each with multiple versions, meshing and integrating multi-supplier technologies to the mix, we add even more complexity, driving up the timescale and cost.
How can we manage this complexity, deliver services with agility and speed, adapt, assure and ensure the economic sustainability of the highly diverse services offered?
Open automated management and orchestration become an urgent necessity to operate networks of such complexity in a profitable way. The ultimate automation target is a largely autonomous network driven by high-level policies and rules. These networks will be capable of self-configuration, self-monitoring, self-healing and self-optimization without further human intervention. These capabilities are enabled by the 3GPP Release 16 specifications.
Enter the standards
3GPP Release 16 provides a powerful and flexible management toolset to cope with the complexities of 5G deployments and service assurance. The Rel-16 specifications (published in June 2020) shift the focus of the management standards from prescribing solutions to introducing a rich, powerful, flexible and interoperable toolset that can easily be used to create innovative and differentiating, automated management and orchestration solutions for 5G radio and core networks and services.
Say goodbye to the inflexible, tightly coupled, vendor-locked proprietary management systems which come with hard-coded data models that are closely dependent on the network element features and software version.
The Rel-16 specifications offer a fundamental set of model-driven management services (defined using the OpenAPI framework), comprehensive data models and rich performance metric definitions that can be used to build modular and scalable management apps and solutions. These combine the power of standardization for multi-vendor plug-and-play integrations with the flexibility of diverse deployment scenarios. The specifications support a wide gamut of features, including 5G radio and core provisioning, network slice management and orchestration (including interactions with verticals and external management systems), data analytics, performance and SLA assurance. All of these features boast automation capabilities.
Operators can simply pick, mix and match management apps (that may come from different vendors/suppliers/open-source) and use them to build innovative, interoperable solutions for rapid 5G network delivery and service assurance, optimized to their special deployment scenario. It is as easy as constructing a Lego model. The specifications are used to fit the Lego blocks together.
Figure 2 depicts an example of how to construct a RAN optimization solution, composed of management apps coming from different suppliers. The solution leverages real-time analytics to continuously optimize RAN behavior and utilization.
In this example, the data-collection app observes and provides live performance and fault management data. The analytics apps predict, detect and correlate events based on that data, providing insight to the intelligence app, which drives the closed-loop automation. The intelligence app decides how to react to an event or observation and triggers the necessary actions by the orchestration or the control apps (e.g. changing the configuration parameters of a network resource, scaling a network slice or adjusting the topology of a network service).
Closed-loop operation enables the continuous optimization and adaption of network-and resource-utilization as well as automated service assurance and fulfilment. Efficient access to real-time management data is the key to analytics, machine intelligence and closed-loop automation. Rel-16 also dispels a common myth about management’s long delay from the time the network data is available to the time appropriate insight can be derived. A new data streaming management service is provided in Rel-16 to support real-time use cases such as real-time investigations, troubleshooting, analytics and optimization. The data-streaming service can be directly consumed in near-real-time by any management app.
Interested in learning more? The Nokia 3GPP SA5 delegates have produced a technical brief describing the 3GPP Rel-16 model-driven management toolset and its features.