Network Performance Root-Cause Analysis via Anomaly Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Modern heterogeneous networks face challenges in optimizing data delivery due to volatility and diversity, leading to inconsistent performance, as existing techniques like caching and compression are ineffective in addressing the dynamic and personalized nature of traffic, and app owners struggle to pinpoint root causes of performance issues without adequate domain expertise or measurement data.
Innovation Solution
A cognitive network performance analysis system that measures data attribute values based on network transactions, tracks anomalies, assesses their impact, and generates alerts to prioritize issues, using a performance analyzer that collects and aggregates metrics from client and server sides to provide informative insights and recommendations for optimizing data delivery.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If compression or right-sizing content techniques are used, then data transmission efficiency is improved, but network volatility and diversity issues remain unaddressed
Solution Approach 1:
The patent segments the network performance analysis into multiple independent components: anomaly detection module, root cause analysis module, and impact assessment module. Each component handles specific aspects of network volatility independently, allowing the system to address network diversity issues while maintaining compression benefits for data transmission efficiency.
Solution Approach 2:
The patent introduces an intermediary performance analysis system that sits between the content delivery network and end users. This intermediary collects metrics data, detects anomalies, and provides root cause analysis without interfering with the compression techniques, thus maintaining transmission efficiency while addressing reliability issues through independent network monitoring.
2Measurement precision
If app owners manually analyze network data to find root causes, then domain expertise can be applied, but time consumption and complexity increase significantly
Solution Approach 1:
The patent implements self-service through automated anomaly detection and root cause analysis capabilities. The system automatically collects metrics data, detects performance anomalies, identifies potential root causes, and assesses their impact without requiring manual domain expertise. This automation maintains high measurement precision while eliminating time-consuming manual analysis.
Solution Approach 2:
The patent establishes a feedback loop where the system continuously monitors network performance metrics, compares them against baseline data, and automatically generates insights about root causes. This closed-loop feedback mechanism enables rapid diagnosis by continuously learning from network behavior patterns, reducing both time and complexity of root cause identification while maintaining accuracy.
3Measurement precision
If comprehensive metrics data is collected from network transactions, then measurement precision is improved, but data complexity and analysis difficulty increase
Solution Approach 1:
The patent extracts only the most relevant metrics data from comprehensive network transaction information. The anomaly detection module selectively collects specific performance metrics rather than processing all available data, reducing analysis complexity while maintaining measurement precision by focusing on key indicators of network performance.
Solution Approach 2:
The patent applies partial action by implementing selective data collection and analysis. Rather than processing all comprehensive metrics data, the system focuses on specific anomaly types and their associated metrics, reducing system complexity while maintaining sufficient measurement precision for effective root cause identification through targeted analysis.
Data Source
AI summary
A data-driven approach to network performance diagnosis and root-cause analysis is presented. By collecting and aggregating data attribute values across multiple components of a content delivery system and comparing against baselines for points of inspection, network performance diagnosis and root-cause analysis may be prioritized based on impact on content delivery. Alerts may be generated to present recommended courses of action based on the tracked performance analysis.