How sifat192/ai-cloud-digital-twin works
This is a Streamlit cloud-resource monitoring dashboard that reads a usage dataset, filters it by user and workload, and presents current metrics, health status, forecasts, anomaly results, visualizations, and optimization recommendations. The principal workflow and recommendation logic are implemented together in app.py; model and scaler artifacts are stored under models/. The sampled code references dataset and model filenames without those directory prefixes, so the graph shows the intended data/model dependencies, not verified runtime path resolution. Other capabilities are based on the README because notebooks were not sampled.
Main entry points: app.py Streamlit dashboard · cpu_forecasting_model.pkl CPU model · cpu_scaler.pkl CPU scaler · memory_forecasting_model.pkl Memory model
Components 16
Dashboard
- Streamlit dashboard
app.py - User/workload filter
app.py - Charts and analysis
app.py
Analytics
- Feature preparation
app.py - Metrics and health
app.py - CPU/memory forecasts
app.py - Anomaly detection
app.py - Optimization recommendations
app.py
Data and models
- Usage dataset
Dataset/cloud_dataset.csv - CPU model
models/cpu_forecasting_model.pkl - CPU scaler
models/cpu_scaler.pkl - Memory model
models/memory_forecasting_model.pkl - Memory scaler
models/memory_scaler.pkl - Anomaly model
models/anomaly_model.pkl - Anomaly scaler
models/anomaly_scaler.pkl
Other components
- Administrator
Connections 24
- Administrator → Streamlit dashboard: uses
- Streamlit dashboard → Usage dataset: reads
- Streamlit dashboard → User/workload filter: renders
- User/workload filter → Feature preparation: selects and encodes
- Feature preparation → Metrics and health: provides records
- Feature preparation → CPU/memory forecasts: provides features
- Feature preparation → Anomaly detection: provides features
- Streamlit dashboard → CPU model: loads
- Streamlit dashboard → CPU scaler: loads
- Streamlit dashboard → Memory model: loads
- Streamlit dashboard → Memory scaler: loads
- Streamlit dashboard → Anomaly model: loads
- Streamlit dashboard → Anomaly scaler: loads
- CPU/memory forecasts → CPU scaler: scales inputs
- CPU/memory forecasts → CPU model: predicts CPU
- CPU/memory forecasts → Memory scaler: scales inputs
- CPU/memory forecasts → Memory model: predicts memory
- Anomaly detection → Anomaly scaler: scales inputs
- Anomaly detection → Anomaly model: predicts anomalies
- Streamlit dashboard → Charts and analysis: renders
- Metrics and health → Optimization recommendations: supplies current metrics
- CPU/memory forecasts → Optimization recommendations: supplies predictions
- Anomaly detection → Optimization recommendations: supplies anomaly status
- Streamlit dashboard → Optimization recommendations: generates