Subscriber-Aware 5G Core Network Traffic Routing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional 5G core network traffic routing methods are inefficient and difficult to scale, lacking subscriber awareness, which prevents effective implementation of service-level policies and leads to inefficient traffic paths due to unnecessary aggregation and disaggregation of traffic.
Innovation Solution
Implementing a segment routing policy that associates subscriber identities with service levels and micro service instances, using SRv6 containers at mobile gateways to route data packets to specific micro service instances based on subscriber identities, allowing for end-to-end monitoring and policy application.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traffic is aggregated and sent to a particular Kubernetes cluster using conventional routing methods, then traffic can be distributed to various nodes, but the routing is inefficient and difficult to scale
Solution Approach 1:
The patent segments the routing process by introducing intermediate routing points (edge devices and UPFs) that perform subscriber identity-based routing decisions. This divides the monolithic core network routing into distributed segments, improving efficiency and reducing complexity at each node.
Solution Approach 2:
The patent performs preliminary routing actions at the edge devices and UPFs by identifying subscriber identities and determining appropriate microservice instances before traffic reaches the Kubernetes cluster. This preliminary segmentation eliminates the need for complex load balancing at the cluster level.
2Adaptability or versatility
If conventional round-robin routing is used to distribute traffic to cluster nodes, then traffic distribution is simple, but subscriber-aware policies cannot be implemented
Solution Approach 1:
The patent applies local quality by implementing subscriber-specific routing policies at distributed edge devices and UPFs. Each routing point can apply different policies based on local subscriber identity information, enabling versatile policy implementation without complex centralized control.
Solution Approach 2:
The patent introduces intermediate routing devices (edge devices and UPFs) that act as mediators between the core network and Kubernetes clusters. These intermediaries perform subscriber identity-based routing decisions, enabling policy implementation while maintaining operational simplicity through automated lookups.
3Productivity
If CPU intensive load balancing schemes are used for traffic disaggregation, then traffic can be distributed based on subscribers, but the system consumes excessive computational resources
Solution Approach 1:
The patent performs preliminary subscriber identity identification and routing decisions at edge devices and UPFs before traffic reaches the Kubernetes cluster. This preliminary action eliminates the need for CPU-intensive load balancing at the cluster level, as routing decisions are already made based on subscriber identities.
Solution Approach 2:
The patent extracts the routing decision-making function from the Kubernetes cluster and places it at edge devices and UPFs. This extraction removes the computational burden of subscriber-aware routing from the cluster, allowing simple IP-based load balancing to handle the actual traffic distribution.
4Productivity
If traffic aggregation is performed before subscriber-based routing, then routing can be centralized, but inefficient traffic paths are created
Solution Approach 1:
The patent segments the network architecture into distributed routing points (edge devices and UPFs) that perform subscriber-based routing before traffic aggregation. This segmentation enables efficient traffic paths by routing traffic closer to its destination based on subscriber identity, avoiding unnecessary backhaul to centralized points.
Solution Approach 2:
The patent performs preliminary subscriber identity-based routing at edge devices and UPFs before traffic is aggregated or forwarded to Kubernetes clusters. This preliminary routing action ensures traffic follows efficient paths from the start, eliminating the need for re-routing later in the network.
Data Source
AI summary
Methods, non-transitory computer readable media, network traffic management devices and network traffic management systems for providing subscriber aware network traffic routing in a 5G network are illustrated. With this technology, a segment routing policy is received from a controller. The segment routing policy includes a plurality of subscriber identities, associated service levels and micro service instances. One or more data packets are received from a user device and the subscriber identity associated with the data packets is identified. Based on the subscriber identity and the segment routing policy, a destination micro service instance is identified and the one or more data packets are forwarded to the destination micro service instance.