SoC & NoC Architecture Simulator

📌 Project Overview

This project provides a high-level modeling and simulation environment for Systems-on-Chip (SoC) and Networks-on-Chip (NoC). Developed in Python, it allows users to architect custom grid-based networks of processing nodes and memory units, and simulate complex data exchange workloads using real-time routing algorithms.

⚙️ Key Features

• Architectural Modeling: Create and configure SoC layouts with dedicated compute and memory nodes.
• Advanced Routing: Implements a Dijkstra-based shortest path algorithm to optimize data packet travel through the NoC.
• Behavioral Simulation: Generate and execute task-based requests to analyze network congestion and latency.
• Data Persistence: Export and import system configurations using Python's pickle format.
• Automated Reporting: Generates detailed simulation logs for post-processing analysis.

🚀 Getting Started

Prerequisites

• Python 3.x installed.
• Standard libraries only (Tkinter for the GUI).

Launching the Interface

To start the graphical user interface, run the following command:

python interface.py

Loading Sample Data

The repository includes pre-configured test files to help you get started quickly:

1. Open NoC: Go to File > Open NoC and select test_noc.pkl.
2. Import Tasks: Go to Tasks > Import Tasks and select test_taches.pkl.
3. Run: Launch the simulation to observe node interactions.

📊 Output

Upon completion, the simulator generates a simulation_report.txt file. This report contains:

• Detailed logs of every data request and transfer.
• Success/Failure status of routing operations.
• Final system state summary.

📁 Repository Structure

• src/: Core logic (Classes: SoC, NoC, Node, Request).
• main.py: Graphical user interface (GUI) implementation.
• data/: Sample .pkl configuration files.

👥 Authors

• Antoine BERTRAND - Main Developer

Samy-William AYYADA - Contributor

🏛 Academic Context

This project was developed in 2023 as part of the Computer Science curriculum at ENSTA Bretagne.