Network Path Latency Scoring for Transient Issue Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network path scoring systems fail to effectively detect transient or ephemeral network problems that can impact application performance and violate service level agreements, particularly in software-defined wide area networks (SD-WANs), due to the dynamic nature of latency and bandwidth utilization.
Innovation Solution
A network path scoring system that calculates a nearly real-time (NRT) health score based on current latency, a non-stationary range of expected/acceptable additional latency, and bandwidth utilization, using a formula that adjusts to load changes and route variations, providing intuitive scoring on a 0 to 100 scale.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional network monitoring methods are used, then network path health can be monitored, but transient or ephemeral network problems cannot be effectively detected
Solution Approach 1:
The patent implements dynamic latency baseline calculation that adapts to changing network conditions. Instead of using fixed thresholds, the system continuously updates expected latency ranges based on historical data and current load conditions, enabling effective detection of transient problems that occur within normal operational variations.
Solution Approach 2:
The system employs feedback mechanisms by continuously monitoring latency measurements and adjusting the baseline expectations accordingly. The latency score is calculated by comparing current latency against dynamically updated baselines, and this feedback loop enables rapid detection of deviations indicating network problems while adapting to legitimate changes in network performance.
2Adaptability or versatility
If latency thresholds are fixed, then scoring is simple, but the system cannot adapt to load changes and route variations
Solution Approach 1:
The patent changes the parameter of latency thresholds from fixed values to dynamic ranges based on load conditions. The system calculates expected latency ranges that vary with network load, allowing the scoring mechanism to adapt to different operational states without requiring complex manual configuration or multiple static threshold sets.
Solution Approach 2:
The scoring system implements dynamic adaptation by continuously updating latency baselines based on observed network behavior under different load conditions. This dynamic approach allows the system to automatically adjust to route changes and load variations without increasing operational complexity for users.
3Measurement precision
If comprehensive network metrics are monitored, then detection accuracy improves, but the scoring system becomes more complex
Solution Approach 1:
The patent extracts and focuses on the most critical metric for network path health: latency. By concentrating on latency measurements and comparing them against dynamically calculated baselines, the system achieves high measurement precision without the complexity of integrating and processing multiple different network metrics simultaneously.
Solution Approach 2:
The scoring system segments the network health assessment into distinct components: baseline latency calculation, current latency measurement, and score computation. This segmentation allows each component to be optimized independently while maintaining overall system simplicity and accuracy.
Data Source
AI summary
A network path scoring system is disclosed herein that scores “health” of network paths based on latency data. The system scores health of a network path based on additional latency of the network path for a current time interval, additional latency expected for the network path, and current load. The scoring uses a non-stationary range that is based on expected additional latency (e.g., engineered/injected latency) and coefficient(s) that vary with load.