How snkas/hypatia works
Main Components
• satgenpy (Python network‑state generator & analysis)
• ns3‑sat‑sim (ns‑3 packet‑level simulator)
• satviz (CesiumJS visualization pipeline)
• paper (experiment orchestration & plotting)
• integration_tests (end‑to‑end test harness)
Relationships & Data Flows
• satgenpy → outputs “network state over time” (static topology + dynamic link state + routing tables)
• ns3‑sat‑sim ← consumes satgenpy outputs → runs ns‑3 simulations → produces raw trace files, logs, gnuplot scripts
• paper ← uses satgenpy outputs & ns3‑sat‑sim results → runs Python scripts + gnuplot → generates publication figures
• satviz ← consumes satgenpy outputs (optionally ns3‑sat‑sim link‑utilization logs) → generates CesiumJS code and static HTML
• integration_tests orchestrate: call satgenpy scripts → call ns3‑sat‑sim build & run → call paper & satviz scripts → verify final outputs
Architectural Patterns & Principles
• Modular pipeline: clear producer/consumer boundaries via file‐based I/O
• Layered abstraction:
• “Network‐state generation” (pure Python)
• “Simulate” (ns‑3 C++ + Python glue)
• “Post‐process & plot” (Python + gnuplot)
• Reuse of open‐source modules: ns‑3 satellite module, basic‑sim, Cesium API
Technologies & External Dependencies
• Python 3.7+, numpy, networkx, astropy, sgp4, exputil, matplotlib, statsmodels, cartopy
• ns‑3 (v3.31) + satellite & basic‑sim modules, OpenMPI, lcov, gnuplot
• Cesium API (requires access token), CesiumJS
• Bash scripts to install deps, build modules, run tests
• Titles: “Hypatia Simulation Framework Architecture”
• Clouds: Python runtime + libs, ns‑3 engine, Cesium API, gnuplot