ML-Based Network Routing for Latency Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current network routing and bandwidth balancing techniques struggle to optimize latency and quality of service in wide area networks, leading to inefficiencies in data transmission.
Innovation Solution
The use of machine learning to classify available routes and determine bandwidth allocations based on historical network routing data, ensuring that routes and bandwidth are optimized to meet defined latency and quality criteria.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If traditional network routing and bandwidth balancing techniques are used, then network infrastructure simplicity is maintained, but latency increases and quality of service deteriorates
Solution Approach 1:
The patent replaces traditional mechanical routing algorithms with machine learning models that classify routes and predict bandwidth allocations. The ML model processes historical network routing data to classify available routes and determine optimal bandwidth allocations, substituting conventional routing mechanisms with intelligent prediction systems that reduce latency and improve quality of service.
Solution Approach 2:
The system employs self-service mechanisms where the machine learning model continuously learns from historical network routing data and autonomously classifies routes and determines bandwidth allocations without requiring complex manual configuration. The model adapts to network conditions automatically, reducing the need for complex administrative interventions while optimizing performance.
2Reliability
If machine learning is applied to classify routes and determine bandwidth allocations, then quality of service improves, but computational requirements and system complexity increase
Solution Approach 1:
The patent implements feedback mechanisms where the machine learning model processes historical network routing data to continuously improve its predictions. The system uses feedback from actual network performance data to refine route classification and bandwidth allocation predictions, enhancing quality of service while managing complexity through data-driven optimization.
Solution Approach 2:
The system performs preliminary classification of available routes and prediction of bandwidth allocations before actual data transmission occurs. By pre-classifying routes and determining optimal bandwidth allocations in advance using historical data, the system reduces real-time computational complexity while improving quality of service through proactive route selection.
3Productivity
If bandwidth allocations are optimized based on historical data, then throughput improves, but data processing requirements increase
Solution Approach 1:
The patent uses copying of historical network routing data to train and refine the machine learning model. By analyzing copies of historical data patterns, the model learns optimal bandwidth allocations and route classifications without requiring processing of every individual data packet in real-time, thus improving throughput while managing data processing requirements through pattern recognition from historical copies.
Data Source
AI summary
The described technology generally relates to modifying network routing and bandwidth balancing for lower latency and high quality fair traffic. For instance, based on network routing data comprising network quality data representative of a network quality applicable to a wide area network and using machine learning applied to historical network routing data, other than the network routing data, available routes are classified, the available routes being between a source node on the wide area network and a destination node on the wide area network. Further, based on the network routing data and using a result of the machine learning applied to the historical network routing data, bandwidth allocations applicable to the classified available routes can be determined. Then, based on the classified available routes and the predicted bandwidth allocations, a route is set for data transmitted from the source node to the destination node.