Knowledge Graph Root Cause Analysis for IT Health Monitoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current business index processing systems fail to establish relationships between key performance indicators (KPIs), making it difficult to diagnose and address technical health issues in IT environments effectively.
Innovation Solution
A system and method that uses a processing unit and evaluator to identify KPIs related to technical health issues, perform root cause analysis, and generate a knowledge graph to discover hidden correlations between KPIs, thereby diagnosing and identifying the root cause of the issue.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional monitoring systems are used to detect technical health issues, then issues can be identified, but it takes a tremendous amount of effort to investigate and determine root causes
Solution Approach 1:
The system performs preliminary root cause analysis by executing queries against the knowledge base before an incident occurs, pre-establishing correlations between KPIs and potential root causes. This preliminary preparation enables rapid diagnosis when issues arise, reducing investigation time while maintaining reliable monitoring.
2Measurement precision
If manual analysis of metrics is performed to understand technical issues, then root causes can be identified, but a large amount of effort is required to build up the pattern of metrics
Solution Approach 1:
The knowledge base performs self-service by automatically maintaining pre-computed correlations between KPIs and potential root causes. The system autonomously executes queries, analyzes patterns, and updates its knowledge without requiring manual intervention, thereby achieving high measurement precision while improving analysis efficiency.
Solution Approach 2:
The system pre-computes and stores correlations between KPIs and potential root causes in the knowledge base before incidents occur. This preliminary action eliminates the need for manual pattern building during actual investigations, maintaining high accuracy while significantly improving productivity.
3Adaptability or versatility
If existing business index processing systems are used, then business performance can be estimated, but they do not establish relationships between KPIs or indices
Solution Approach 1:
The knowledge base is designed with multi-functionality to handle both business performance estimation and KPI relationship analysis. It universally processes various types of data including KPIs, metrics, and their correlations, enabling the system to maintain adaptability while recovering the lost KPI correlation information through comprehensive relationship mapping.
Data Source
AI summary
Embodiments relate to monitoring an information technology (IT) environment having a plurality of domains through key performance indicator (KPI) data. In response to detection of a technical health problem, a first KPI related to the problem is identified. A root cause analysis is performed on the identified KPI generating a knowledge graph. A second KPI related to the first KPI is identified through the discovery of a correlation between the two identified KPIs. A diagnosis is generated for the technical health problem within the IT environment based on the discovered hidden correlation between the first KPI and second KPI. The generated diagnosis includes the root cause of the technical health issue.