Setting File Incident Analysis via Action Record Pattern Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In complex systems with multiple apparatuses, determining the specific setting items causing incidents due to configuration errors is challenging, especially when multiple middleware products are involved, as existing methods fail to accurately associate incidents with the responsible setting items.
Innovation Solution
An apparatus analyzes action records to extract setting-item names, counts their occurrences, and generates occurrence patterns, comparing them to model patterns to identify setting files that may cause incidents, thereby determining causal items and associating them with incident information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple middleware products are combined in a layered manner, then system functionality and versatility are improved, but system complexity and difficulty in determining failure causes increase
Solution Approach 1:
The patent segments the complex system into multiple layers (infrastructure layer, middleware layer, application layer) and further segments failure analysis into distinct steps: obtaining action records, extracting setting-item names, counting occurrences, generating occurrence patterns, comparing with model patterns, and determining causal setting files. This segmentation allows each layer and each analysis step to be managed independently, reducing the complexity of analyzing the entire system.
Solution Approach 2:
The patent introduces an intermediary analysis mechanism that uses action records as intermediate data between system operations and failure diagnosis. The occurrence pattern comparison acts as an intermediary process that bridges the gap between complex multi-layer system behavior and simple cause identification, enabling automated failure analysis without requiring direct inspection of all system components.
2Measurement precision
If comprehensive action records are analyzed to identify setting errors, then measurement precision of failure causes is improved, but loss of time for analysis increases
Solution Approach 1:
The patent performs preliminary actions by pre-obtaining and storing model patterns of setting errors from historical data before actual failure analysis is needed. When a failure occurs, the system directly compares occurrence patterns against these pre-prepared model patterns, eliminating the need for time-consuming real-time analysis of all possible causes. This preliminary preparation significantly reduces analysis time while maintaining high precision.
Solution Approach 2:
The patent replaces manual, mechanical analysis of action records with an automated computational system. The system automatically extracts setting-item names, counts occurrences, generates occurrence patterns, and compares them with model patterns using algorithmic processes. This substitution of automated information processing for manual analysis dramatically reduces analysis time while improving measurement precision through consistent, error-free computation.
3Reliability
If setting-item occurrence patterns are compared with model patterns, then reliability of incident analysis is improved, but device complexity of the analysis apparatus increases
Solution Approach 1:
The patent creates simplified copies of setting files in the form of occurrence patterns and model patterns. Instead of analyzing the complete, complex setting files and their entire context, the system works with condensed copies that capture only the essential characteristics needed for failure diagnosis. These pattern copies retain the critical information about setting-item occurrences while eliminating unnecessary complexity, enabling reliable comparison with minimal computational overhead.
Data Source
AI summary
For each setting-file, an apparatus obtains, from first action-records, an occurrence-count value that is associated with the each first action-record and indicates a number of occurrences of a setting-item name identifying each setting-item for the each setting-file, in association with a sequence number assigned to the each first action-record, where each first action-record records an action that is taken in response to a first-incident. When there exists a first setting-file for which an occurrence-pattern generated based on a sequence of occurrence-count values that are each associated with different one of the sequence numbers assigned to the first action-records is similar to a model pattern that is in advance obtained from second action-records for second-incidents occurring due to a setting error of a setting-value that is set to the first setting file, the apparatus stores the first setting-file in association with an identifier identifying the first-incident.