How rajinikanthvadla-ai/aiops-realworld-labs works
This AIOps lab collects historical CPU metrics from Prometheus, trains an Isolation Forest model, then repeatedly checks live CPU readings and sends Slack alerts for anomalies. The scripts and their artifact handoffs are shown as the main workflow; configuration and external services are included where visible. The README gives no further product description, and the deployment/configuration files were not sampled, so their wiring is not asserted.
Main entry points: detect_anomalies.py Detect Anomalies · fetch_metrics.py Fetch Metrics · train_model.py Train Model · config.ini Monitoring Config
Components 9
Metric Collection
- Fetch Metrics
DAY-01/aiops-lab/scripts/fetch_metrics.py - Monitoring Config
DAY-01/aiops-lab/scripts/config.ini - CPU Metrics CSV
Model Training
- Train Model
DAY-01/aiops-lab/scripts/train_model.py - Trained Model
Anomaly Monitoring
- Detect Anomalies
DAY-01/aiops-lab/scripts/detect_anomalies.py
Other components
- Operator
- Prometheus
- Slack
Connections 12
- Operator → Fetch Metrics: runs
DAY-01/aiops-lab/scripts/fetch_metrics.py - Operator → Train Model: runs
DAY-01/aiops-lab/scripts/train_model.py - Operator → Detect Anomalies: runs
DAY-01/aiops-lab/scripts/detect_anomalies.py - Fetch Metrics → Prometheus: queries history
DAY-01/aiops-lab/scripts/fetch_metrics.py - Fetch Metrics → Monitoring Config: reads
DAY-01/aiops-lab/scripts/fetch_metrics.py - Fetch Metrics → CPU Metrics CSV: writes
DAY-01/aiops-lab/scripts/fetch_metrics.py - Train Model → CPU Metrics CSV: reads
DAY-01/aiops-lab/scripts/train_model.py - Train Model → Trained Model: writes
DAY-01/aiops-lab/scripts/train_model.py - Detect Anomalies → Trained Model: loads
DAY-01/aiops-lab/scripts/detect_anomalies.py - Detect Anomalies → Monitoring Config: reads
DAY-01/aiops-lab/scripts/detect_anomalies.py - Detect Anomalies → Prometheus: queries live data
DAY-01/aiops-lab/scripts/detect_anomalies.py - Detect Anomalies → Slack: posts alerts
DAY-01/aiops-lab/scripts/detect_anomalies.py