Root Cause Analysis System for Network Event Diagnostics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Network operators face significant challenges in performing root cause analysis for complex network events, as existing tools are often specific to certain tasks, lack automated rule learning capabilities, and require extensive manual data gathering, making it impractical for large networks to quickly detect and resolve faults and performance issues.
Innovation Solution
A root cause analysis system that includes a data gatherer, join finder, and root cause identifier, which retrieves symptom event instances, generates diagnostic events based on dependency rules, and identifies root causes using rule-based reasoning and Bayesian inference, enabling automated analysis and generation of trouble tickets.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If existing root cause analysis tools are used, then analysis can be performed for specific tasks, but the tools lack automated rule learning capabilities and require extensive manual data gathering
Solution Approach 1:
The system enables automated rule learning where the root cause analysis system automatically learns and updates its own diagnostic rules from historical network event data without requiring manual programming. The system serves itself by autonomously improving its analytical capabilities through machine learning algorithms that process past events and generate updated dependency rules.
Solution Approach 2:
The patent replaces manual mechanical data gathering and rule creation processes with automated computational systems. Machine learning algorithms automatically learn diagnostic rules from data, substituting the manual expertise and labor previously required for rule development and system configuration.
2Productivity
If manual data gathering is performed, then comprehensive analysis can be achieved, but the process requires extensive time and effort
Solution Approach 1:
The system performs preliminary actions by pre-processing and normalizing network event data as it is collected, preparing it for analysis in advance. Historical data is continuously preprocessed and stored in standardized formats, so when analysis is needed, the data is already ready for immediate querying and evaluation, eliminating the need for time-consuming manual data gathering at the time of incident.
Solution Approach 2:
Manual data gathering operations are replaced with automated data collection and processing systems that continuously monitor network devices, automatically retrieve events, and prepare data for analysis without human intervention, dramatically reducing the time required for data acquisition.
3Adaptability or versatility
If existing tools are used, then specific tasks can be addressed, but they are not practical for large networks with diverse network events
Solution Approach 1:
The root cause analysis system is designed with universal applicability to handle diverse network events across multiple network devices and protocols. The system uses standardized data normalization and generic diagnostic frameworks that can adapt to various network configurations and event types, making it applicable to large, heterogeneous networks without requiring task-specific customization.
Solution Approach 2:
Manual analysis operations are replaced with automated diagnostic systems that can independently analyze diverse network events using learned rules and algorithms, making the system easy to operate without requiring expert knowledge for each specific network scenario.
Data Source
AI summary
Example methods, apparatus and articles of manufacture to perform root cause analysis for network events are disclosed. An example method includes retrieving a symptom event instance from a normalized set of data sources based on a symptom event definition; generating a set of diagnostic events from the normalized set of data sources which potentially cause the symptom event instance, the diagnostic events being determined based on dependency rules; and analyzing the set of diagnostic events to select a root cause event based on root cause rules.