Distributed GTP-C Processing for Low-Delay Routing

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Users have no control over which gateway is used to connect to the public network, leading to potential long exposure of data in the public internet when the break-out point is far from the data target, such as an application server, resulting in increased delays and reduced quality of service.

Innovation Solution

A virtual core network with clusters of gateways, load balancers, and subscriber databases allows users to choose the preferred connection point, routing traffic through a high-quality, low-delay private network to minimize exposure to the public internet by selecting the least loaded gateway based on user preferences and load criteria.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If the telecommunication network routes traffic through a central node to the enterprise, then the network can collect and manage traffic efficiently, but the data exposure to the public internet increases and delays increase

Engineering Contradiction:
Improvetraffic management efficiencyVSAvoiddata exposure to public internet
Core Design Contradiction:
ProductivityVSObject-affected harmful factors

Solution Approach 1:

The network is segmented into multiple distributed GTP-C processing nodes located at different network edges and enterprise proximity points, rather than a single central node. Each node independently processes routing decisions, enabling traffic to be routed closer to destination enterprises and reducing public internet exposure while maintaining efficient traffic management through distributed coordination

Inventive Principle:
Principle #1Segmentation

2Device complexity

If the telecommunication network uses a single gateway for connection, then the network structure is simple, but the user has no control over the connection point and experiences longer delays

Engineering Contradiction:
Improvenetwork structure complexityVSAvoidconnection delay
Core Design Contradiction:
Device complexityVSLoss of time

Solution Approach 1:

The network transitions from a single-dimension centralised architecture to a multi-dimensional distributed architecture with GTP-C processing nodes positioned at multiple network edges and enterprise proximity points. This spatial distribution across multiple dimensions enables users to connect to the nearest optimal gateway, reducing connection delays while the modular design maintains manageable structural complexity

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

3Ease of manufacture

If the network routes traffic over the public internet to reach application servers, then the routing is simple, but the quality of service decreases and delays increase

Engineering Contradiction:
Improverouting implementation simplicityVSAvoidquality of service
Core Design Contradiction:
Ease of manufactureVSReliability

Solution Approach 1:

Distributed GTP-C processing nodes act as intermediary elements between the core network and enterprise networks. These intermediaries establish direct private network connections to enterprises, bypassing the public internet for critical traffic while maintaining simple routing protocols. The intermediaries enable high-quality service through dedicated connections without significantly increasing routing complexity

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentUS11310847B2Multi-layered distributed GTP-C processing
Publication Date: 2022.04.19 EMNIFY GMBH
  • US11310847B2 patent drawing
  • US11310847B2 patent drawing
  • US11310847B2 patent drawing

AI summary

A method for multi-layered distributed GTP-processing for communication between user devices (2, 3) via a communication network (1), the network comprises clusters (4) each cluster (4) including gateways (7) and a loadbalancer (8), and a runtime database (5) storing load information of the gateways and a subscriber data-base (6) storing information on user preferences. The method comprises the loadbalancer (8), after receipt of a connection request, to choose the least loaded gateway (7), identifying if the cluster (4) is the preferred cluster of the user, and establishing, if the cluster is the preferred cluster, a connection between the user devices (2, 3) via the gateway (7) and in accordance with the user preferences or else, transmitting the connection request to the loadbalancer (8) of the preferred cluster (4), the loadbalancer identifying the least loaded gateway (7) in that cluster and establishing via the gateway (7) a connection between user devices by sending a response message in accordance with the user preferences to the first gateway in the first cluster.