What is Anomaly Detection?
Scalelite Manager Pro's anomaly detection uses machine learning algorithms to automatically identify unusual patterns in your BigBlueButton infrastructure. Instead of manually setting thresholds, the system learns what "normal" looks like for your specific environment and alerts you when something deviates from that baseline.
How Anomaly Detection Works
Collect Data
→
Learn Baseline
→
Detect Deviation
→
Alert & Act
Key Capabilities
Time-Based Learning
The system understands that Monday mornings look different from Friday evenings, automatically adjusting expectations based on time patterns.
Multi-Metric Analysis
Correlates CPU, memory, network, and meeting patterns to detect issues that single-metric monitoring would miss.
False Positive Reduction
Machine learning reduces alert fatigue by distinguishing between real problems and normal fluctuations.
Early Warning
Detects subtle trends before they become critical issues, giving you time to respond proactively.
Learning Period
The anomaly detection system requires approximately 7 days of data collection to establish accurate baselines for your infrastructure. During this period, you'll still receive threshold-based alerts.
Types of Anomalies Detected
- Resource Anomalies: Unusual CPU, memory, or disk usage patterns
- Traffic Anomalies: Unexpected spikes or drops in meeting activity
- Performance Anomalies: Degraded response times or quality metrics
- Behavioral Anomalies: Unusual meeting patterns or participant behavior
- Infrastructure Anomalies: Server health deviations or network issues
Next Steps
Now that you understand how anomaly detection works, learn how to configure sensitivity settings for your specific needs.