Over the past decade, the telecommunication industry has undergone a profound transformation, witness-ing substantial shifts in the design, deployment, and operation of network infrastructure. The advent of 5G technology has set forth a new set of requirements and challenges for network operators. Predominantly, these challenges are accentuated at larger scales of the network and the encompassed data, which are the core facilitators for emergent use cases foreseen in the Beyond 5G and 6G. Despite this evolving landscape, the prime objective for network operators remains to be the same: To optimize the cost-revenue problem, al-beit in a substantially expanded dimension. This demands a transformation from the traditional, manual network management approaches to the modern, and automated, cloud-native systems. This transforma-tion has previously occurred with Self-Organizing Networks (SON) falling short in addressing the challenges of 5G networks [1]. Now, with the surge of cloud-native technologies into the telecommunication domain, this transformation has to be elevated further.
In the context of the cost-revenue optimization, a pivotal factor in network operation costs is resource con-sumption, whereas revenue generation is primarily influenced by sustained quality of service and business agility for a rapid service introduction. In the realm of cloud-native 5G and 6G, the former encompasses concepts such as cost and energy optimization, resource sharing, and network slicing, while the latter lever-ages a majority of DevOps practices, relatively new to the telecommunication industry. The advent of cloud computing and virtualization technologies has empowered network operators to minimize Capital and Oper-ational Expenditures by consolidating Network Functions (NFs) onto commodity hardware. Advancements in cloud technologies have facilitated the instantiation and decommissioning of NFs, shifting network man-agement from lengthy, manual processes to agile, automated Management and Operation (MANO) systems. This paradigm shift introduces a proactive network design strategy aimed at meeting service requirements, rather than retrofitting services to fit a pre-established network architecture. It has also redefined problem statements such as network slicing, transitioning from a resource division focus to a design approach aligned with slice requirements and available resources. These dual formulations fundamentally alter how MANO systems are designed and operated for the dynamic, software-oriented, cloud-native networks, necessitating a new generation of MANO that adheres to the cloud-native principles like declarative reconciliation and consistent automation.
On the other hand, both traditional and current MANO systems have lagged behind the rapid advancements in the IT industry, failing to fully exploit the capabilities of cloud-native environments. To bridge this gap, this thesis introduces a specialized cloud-native MANO for next-generation mobile networks, conceptu-alized under the TRIREMATICS project, which embodies cloud-native principles to redefine and revamp the MANO stack fundamentally. Our newly proposed methods significantly outperform existing solutions in all considered metrics, such as agility and overhead. Furthermore, a substantial number of operations become scalable only when these modern methodologies are integrated.
