Paul Baran Network Topologies Simulation

This project reproduces the results of Paul Baran's foundational research on network robustness using a Monte-Carlo simulation approach. It was developed as part of a lab for the Architecture of Computer Networks course at HEPIA.

---

Summary

• Simulates the impact of random node failures on different network topologies.
• Measures the resilience of topologies with increasing redundancy levels (R = 1 to 6).
• Uses statistical averaging and visual plots to observe survival-rate trends.

---

Context

Paul Baran—one of the pioneers of packet‑switched networks—showed that distributed networks are far more resilient than centralized ones. This project replicates his findings by:

• Building grids with various redundancy levels R
• Randomly removing nodes (failures)
• Measuring the largest connected component (LCC) after each failure
• Plotting robustness vs. redundancy

---

Technologies Used

• Python 3.10+
• NetworkX – graph creation & analysis
• NumPy – random sampling
• Matplotlib – result visualisation

---

Getting Started

# 1. Clone the repository
git clone https://github.com/synloop/paul-baran-network-topo.git
cd paul-baran-network-topo

# 2. Install dependencies
pip install -r requirements.txt

# 3. Run the simulation
python simulation.py

The script generates graphs showing survival rates as node‑failure probability increases.

---

Project Structure

.
├── simulation.py          # Monte‑Carlo simulation driver
├── print_topology.py      # Helper – visualise a given topology
├── test_breakdowns.py     # Extra script for custom experiments
├── requirements.txt
├── README.md
└── docs/
    └── report.pdf         # Full lab report (French)

Read the full lab report (PDF, in French)

---

Sample Results

• R = 1 (Line topology) : fails quickly under random removals
• R = 2 → R = 4 : drastic improvement in resilience
• R = 5 & R = 6 : diminishing returns despite extra complexity

Plots include error bars (standard deviation) computed over 100+ simulations.

---

What I Learned

• Network resilience matters: Paul Baran’s 1964 work showed that redundancy allows distributed networks to survive random failures — an idea I reproduced through simulation.
• Python graph tooling: Using NetworkX gave me hands-on experience building, analyzing, and visualizing large network graphs programmatically.
• Monte Carlo thinking: Running many random failure scenarios helped me understand the role of statistical methods in stress-testing network architectures.

---

License

Released under the MIT License. See the LICENSE file for details.

📝 Final grade received for this project: 5.8 / 6.0 (Swiss grading system)