Root Cause Localization via Topological Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing root cause analysis methods for cyber-physical systems are costly and error-prone due to the complexity of these systems and the vast amount of monitoring data, often failing to capture the intricacies of abnormal patterns associated with system failures.
Innovation Solution
A method that combines system logs and system metrics into time-series data, performs individual and topological root cause analysis, integrates causal scores and topological patterns using a weighted sum, and executes corrective actions on identified entities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional root cause analysis methods are used on cyber-physical systems, then comprehensive analysis can be performed, but the analysis is costly and error-prone due to system complexity and vast monitoring data
Solution Approach 1:
The patent segments the root cause analysis into two distinct components: individual root cause analysis (examining each system entity separately) and topological root cause analysis (examining relationships between entities). This segmentation allows the complex analysis task to be divided into manageable parts, reducing errors while maintaining comprehensive coverage of the cyber-physical system.
Solution Approach 2:
The patent introduces a topological dimension to traditional root cause analysis by analyzing the spatial and relational structures among system entities. By adding this dimensional aspect (topological patterns) to the conventional individual entity analysis, the method captures intricate abnormal patterns that single-dimension approaches miss, thereby improving accuracy without being overwhelmed by system complexity.
2Reliability
If comprehensive monitoring data is collected from all system entities, then complete root cause analysis is possible, but the time to isolate problems increases
Solution Approach 1:
By segmenting the analysis into individual and topological components, the patent enables parallel processing of different data aspects. This segmentation allows the system to process comprehensive monitoring data more efficiently, reducing the time required to isolate problems while maintaining complete and reliable analysis coverage.
Solution Approach 2:
The patent applies partial action by focusing computational resources on the most relevant aspects of the data through the two-stage analysis approach. Instead of uniformly processing all data with equal depth, the method selectively applies individual analysis to specific entities and topological analysis to relevant relationships, thereby maintaining analysis completeness while reducing overall processing time.
3Productivity
If single-modality root cause analysis is performed, then analysis speed is maintained, but the intricacies of abnormal patterns are not captured
Solution Approach 1:
The patent merges two distinct analysis modalities—individual root cause analysis and topological root cause analysis—into a unified framework. This combination allows the system to maintain analysis speed through efficient individual entity examination while simultaneously capturing intricate abnormal patterns through topological relationship analysis, thereby achieving both productivity and measurement precision.
Data Source
AI summary
Methods and systems for root cause analysis include combining system logs and system metrics into time-series data. Individual root cause analysis is performed to determine individual causal scores for respective system entities. Topological root cause analysis is performed to capture topological patterns of system anomalies. The individual causal scores and the topological patterns are integrated by a weighted sum. A corrective action is performed on an entity identified based on the weighted sum.