An Optimization-Based Orchestrator for Resource Access and Operation Management in Sliced 5G Core Networks

Area of Science:

• Telecommunications Engineering
• Computer Networking
• Operations Research

Background:

• Network slicing allows deploying services with heterogeneous Quality of Service (QoS) requirements.
• Efficient slice resource provisioning and network operation require a sophisticated orchestrator.
• Internet Service Providers (ISPs) need sustainable network evolution considering scalability and flexibility.

Purpose of the Study:

• To design an orchestrator for network slicing focused on efficient resource provisioning for ISPs.
• To analyze critical factors: access control, scheduling, and resource migration for sustainable networks.
• To formulate and solve the resource management problem to satisfy high QoS applications.

Main Methods:

• Formulation of the resource management problem as a mixed-integer programming (MIP) problem.
• Development of a solution approach based on Lagrangian relaxation (LR) for the orchestrator.
• Modeling mechanisms in a pay-as-you-go manner for high system utilization.

Main Results:

• The proposed orchestrator effectively investigates resources for access control within a cost-efficient pool.
• Efficient allocation and migration of resources across network slices are achieved.
• Experimental results demonstrate near-optimal system revenue while meeting QoS requirements.

Conclusions:

• The developed orchestrator provides an efficient solution for network slicing resource management.
• The approach ensures high QoS satisfaction and optimizes system revenue for ISPs.
• The pay-as-you-go model enhances system utilization and sustainability.