Cognitive Debugging System for Automated Fault Diagnosis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Managed system log file analysis is a manual, resource-intensive process that requires skilled individuals to diagnose faults, often necessitating multiple experts due to the disparate nature of components and limited strategic use of debug mode.
Innovation Solution
A cognitive debugging method and system that extracts context from events, maps it to relevant components, enables debug mode, generates log entries, and matches them to pre-stored entries to provide problem resolution documents, reducing manual intervention and resource consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual log file analysis is performed by skilled individuals, then fault diagnosis accuracy is improved, but resource consumption and time requirements increase
Solution Approach 1:
The system performs preliminary actions by pre-processing log files, pre-classifying anomalies, and pre-loading diagnostic rules into the cognitive system before actual fault events occur. This preparation enables rapid automated analysis when faults occur, maintaining high diagnostic accuracy without requiring manual intervention time.
Solution Approach 2:
The cognitive system creates and maintains a copy of expert diagnostic knowledge, rules, and patterns in an automated system. This digital copy of expert knowledge enables the system to perform diagnostics independently, replicating expert-level accuracy while eliminating the time cost of manual expert analysis.
2Loss of information
If debug mode is invoked in selected components, then detailed diagnostic information is obtained, but system resource consumption increases
Solution Approach 1:
The system applies local quality by enabling debug mode and detailed logging only in specific components where anomalies are detected, rather than system-wide. The cognitive system analyzes log patterns to identify which particular components require deep diagnostic scrutiny, concentrating computational and system resources only where needed.
Solution Approach 2:
The system employs partial action by selectively invoking debug mode in only those components showing anomalous behavior, rather than enabling it across all components. This partial application of debug mode sufficient to diagnose the specific anomaly without the excessive resource consumption of universal debug activation.
3Adaptability or versatility
If multiple experts are deployed to diagnose disparate components, then comprehensive fault coverage is improved, but operational complexity increases
Solution Approach 1:
The cognitive system achieves universality by designing a single multi-functional platform that can diagnose all types of components (storage, computing, networking, etc.) through a unified anomaly detection and analysis engine. This universal system replaces the need for multiple specialized experts, covering diverse fault types while simplifying operations through consistent automated processes.
Solution Approach 2:
The system merges the capabilities of multiple expert diagnosticians into a single integrated cognitive system. By combining anomaly detection, log analysis, pattern recognition, and diagnostic rule evaluation into one unified automated platform, the system achieves comprehensive fault coverage that previously required multiple human experts while dramatically reducing operational complexity.
Data Source
AI summary
A method for the cognitive debugging of a managed system includes first receiving an event in an event management system. Thereafter, a context for the event is extracted therefrom and the context is mapped to both one or more components of a managed computing system and also one or more corresponding debug mode commands for each of the components. Consequently, a debug mode is enabled in each of the components and the corresponding debug mode commands are issued for each of the components so as to provoke a generation of one or more log entries. The generated log entries then are matched to a pre-stored log entry amongst a multiplicity of pre-stored log entries and at least one problem resolution document stored in connection with the matched pre-stored log entry is transmitted to an operator of the event management system.