Dynamic Baseline Adjustment for IT Metric Monitoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Configuration changes in IT systems often lead to false positive alerts or obscured real problems due to metrics behaving differently, as existing performance monitoring systems fail to accurately adjust baselines in response to such changes.
Innovation Solution
A mechanism is introduced to automatically adjust the baseline of metrics in IT systems by detecting configuration changes, identifying linkages between configuration parameters and metrics, and applying baseline adjusting algorithms to rescale or adapt the baselines, thereby eliminating false alarms and capturing real issues.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the baseline is kept fixed after configuration change, then the system can detect real problems that deviate from normal behavior, but false positive alerts are generated when metrics change due to configuration rather than actual problems
Solution Approach 1:
The patent applies the Dynamics principle by making the baseline dynamic rather than fixed. When a configuration change is detected, the system automatically adjusts the baseline to reflect the new normal behavior caused by the configuration change. This resolves the contradiction by allowing the baseline to adapt (improving reliability) while preventing false positives through automatic adjustment.
Solution Approach 2:
The patent applies the Parameter changes principle by modifying the baseline parameter in response to configuration changes. The system detects configuration parameters, identifies their impact on metrics, and adjusts the baseline parameters accordingly. This allows the system to maintain accurate detection (improving reliability) while eliminating false alerts caused by configuration-induced metric changes.
2Object-generated harmful factors
If the baseline is adjusted frequently to adapt to configuration changes, then false positive alerts are reduced, but the complexity of the monitoring system increases due to baseline adjustment algorithms and configuration tracking
Solution Approach 1:
The patent applies the Preliminary action principle by pre-establishing the relationship between configuration parameters and metrics before configuration changes occur. The system maintains a knowledge base or mapping that identifies which metrics are affected by which configuration parameters. When a configuration change is detected, the system can immediately adjust the appropriate baselines without complex real-time analysis, thus reducing false positives while keeping system complexity manageable.
Solution Approach 2:
The patent applies the Feedback principle by implementing a closed-loop system that monitors configuration changes and automatically adjusts baselines in response. The system detects configuration parameter changes, identifies affected metrics through pre-established relationships, adjusts baselines accordingly, and continues monitoring. This automated feedback loop reduces false positives while managing complexity through systematic rather than ad-hoc adjustments.
Data Source
AI summary
A baseline adjusting technique allows automatically adjust the baselines of metrics affected by a configuration change a monitored system. If a configuration change is detected, a performance management system retrieves linkages between the changed configuration parameter and one or more metrics. The performance management system then adjusts the baselines of the metric using the baseline adjusting algorithm retrieved from the linkage.