Data Center Issue Impact Analysis Using AI Ops
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Data centers face challenges in analyzing and prioritizing data center issues due to the complexity of interdependencies among assets and the frequent occurrence of both minor and severe issues, which can lead to service degradation and significant business impact, especially in large environments where timely remediation is critical.
Innovation Solution
A system and method for data center monitoring and management that utilizes AI Ops combining big data and machine learning to perform data center event correlation, anomaly detection, and causality determination, enabling contextualization, issue analysis, and prioritization based on asset alert data to identify affected assets and determine the severity and business impact of issues.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data center monitoring covers all assets and events, then completeness of issue detection is improved, but system complexity and processing burden increase
Solution Approach 1:
The patent segments the data center monitoring system into multiple functional components: asset monitoring module, event correlation module, impact analysis module, and prioritization module. Each component handles specific aspects of monitoring, allowing the system to process information in manageable segments rather than attempting to analyze all data simultaneously, thus reducing overall system complexity while maintaining comprehensive coverage.
Solution Approach 2:
The patent introduces an event correlation module that acts as an intermediary between raw event data and final impact analysis. This intermediary layer processes and correlates events from multiple assets, filtering out noise and identifying patterns before the impact analysis stage, thereby reducing the processing burden on subsequent components while maintaining detection completeness.
2Measurement precision
If the system analyzes all data center events in detail, then issue prioritization accuracy is improved, but processing time and response speed decrease
Solution Approach 1:
The patent performs preliminary actions by pre-establishing asset dependency relationships and impact models before actual issues occur. The system pre-maps how failures in one asset affect other assets, so when an issue is detected, the impact analysis can be performed quickly using pre-computed relationships rather than analyzing all possible scenarios in real-time, thus maintaining accuracy while improving response speed.
Solution Approach 2:
The patent applies local quality by tailoring the level of analysis to the specific context of each issue. Not all events require the same level of detailed analysis; the system adjusts the depth of impact analysis based on the severity, type, and context of each event, allocating processing resources more efficiently and reducing overall processing time while maintaining prioritization accuracy for critical issues.
3Measurement precision
If the system tracks all asset interdependencies, then accuracy of impact assessment is improved, but data processing volume and complexity increase
Solution Approach 1:
The patent extracts only the relevant dependency information needed for impact assessment from the complete asset interdependency graph. Instead of processing all possible asset relationships, the system extracts and focuses on dependencies relevant to the specific issue being analyzed, reducing data processing volume while maintaining impact assessment accuracy for the affected areas.
Solution Approach 2:
The patent implements dynamic impact analysis where the scope of dependency tracking adjusts based on the current state and context. The system dynamically determines which asset relationships are relevant to a given issue based on current operational status, asset criticality, and issue severity, thereby processing only the necessary data volume required for accurate impact assessment at any given time.
Data Source
AI summary
A system, method, and computer-readable medium are disclosed for performing a data center monitoring and management operation. The data center monitoring and management operation includes: monitoring a plurality of data center assets contained within a data center; receiving a data center asset data stream, the data center asset data stream comprising a plurality of data center events, at least some of the plurality of data center events having associated data center issues; capturing asset alert data from the plurality of data center events; and, performing a data center issue analysis operation, the data center issue analysis operation using the asset alert data from the plurality of data center events, the data center issue analysis operation identifying a data center asset from the plurality of data center assets affected by the particular data center issue.