Distributed CPU Monte Carlo Traffic

Grid Computing Simulation Platform

---

📋 Table of Contents

1. Project Overview
2. System Architecture
3. How It Works
4. Technologies Used
5. Installation & Execution
6. Network Configuration
7. Scalability & Multi-Workers
8. Future Improvements
9. Contact

---

🚀 Project Overview

Distributed CPU Monte Carlo Traffic is a high-performance Grid Computing platform designed to distribute CPU-intensive Monte Carlo simulations across multiple machines (Workers) managed by a central Master node.

🎯 Goals

• Leverage distributed CPU power across multiple nodes
• Execute simulations in parallel for faster computation
• Aggregate results efficiently with minimal overhead
• Provide a modern user interface using Next.js for visualization
• Support dynamic worker scaling for on-demand computation

---

🏗️ System Architecture

graph TB
    subgraph "Frontend Layer"
        F[Next.js Client]
    end
    
    subgraph "Coordination Layer"
        R[RMI Registry]
    end
    
    subgraph "Control Layer"
        M[Master Node]
    end
    
    subgraph "Computation Layer"
        W1[Worker Node 1]
        W2[Worker Node 2]
        W3[Worker Node N...]
    end
    
    F -->|Start Simulation| M
    R -->|Service Discovery| M
    R -->|Service Registration| W1
    R -->|Service Registration| W2
    R -->|Service Registration| W3
    M -->|Distribute Tasks| W1
    M -->|Distribute Tasks| W2
    M -->|Distribute Tasks| W3
    W1 -->|Return Results| M
    W2 -->|Return Results| M
    W3 -->|Return Results| M
    M -->|Aggregate & Display| F

Loading

Architecture Components

Component	Role	Description
RMI Registry	Service Discovery	Central registry for coordinating nodes
Master Node	Task Manager	Splits simulations, assigns tasks, aggregates results
Worker Nodes	Computation Units	Perform CPU-intensive calculations
Client/Frontend	User Interface	Next.js app for simulation control and visualization

---

⚙️ How It Works

Workflow Overview

sequenceDiagram
    participant C as Client
    participant R as RMI Registry
    participant M as Master
    participant W1 as Worker 1
    participant W2 as Worker 2
    
    Note over R: 1. Start RMI Registry
    Note over M: 2. Master registers with Registry
    Note over W1,W2: 3. Workers register with Registry
    
    C->>M: 4. Launch Simulation
    M->>M: 5. Split into Tasks
    M->>W1: 6. Assign Task 1
    M->>W2: 7. Assign Task 2
    par Parallel Execution
        W1->>W1: Compute Task 1
        W2->>W2: Compute Task 2
    end
    W1->>M: 8. Return Results
    W2->>M: 9. Return Results
    M->>M: 10. Aggregate Results
    M->>C: 11. Display Final Results

Loading

Key Features

• Asynchronous execution - Workers process tasks independently
• Dynamic scaling - Workers can join/leave during execution
• Fault tolerance - System continues if workers disconnect
• Load balancing - Tasks distributed based on worker availability

---

🛠️ Technologies Used

Backend (Grid Computing)

Frontend

Tools & Environment

---

---

🚀 Installation & Execution

1. Clone the Repository

git clone https://github.com/ZakariaHanani/distributed-cpu-montecarlo-traffic.git
cd distributed-cpu-montecarlo-traffic

2. Build the Project

mvn clean install

3. Start the RMI Registry

./scripts/run-registry.sh

Output: ✅ RMI Registry started on port 1099

4. Start the Master Node

./scripts/run-master.sh

Output: ✅ Master registered successfully

5. Configure and Start Workers

⚠️ Each worker must define its own machine IP address

Edit scripts/run-worker.sh:

WORKER_IP=192.168.1.10  # Replace with your machine IP

Start the worker:

./scripts/run-worker.sh

Output: ✅ Worker registered: [email protected]

6. Launch the Client

./scripts/run-client.sh

7. Access the Web Interface

cd frontend
npm run dev

Visit: http://localhost:3000

---

🌐 Network Configuration

Worker Configuration

Each worker must:

• Use its local machine IP (not localhost)
• Be reachable over the network
• Have appropriate firewall rules

Central Configuration

The Master and Registry can load network settings from:

# config.properties
registry.host=192.168.1.1
registry.port=1099
master.host=192.168.1.2
worker.pool.size=10

Benefits:

• Single configuration file
• No code changes for network adjustments
• Easy deployment across environments

---

📈 Scalability & Multi-Workers

Dynamic Scaling

┌─────────────────────────────────────┐
│          Master Node                │
│  ┌────────────────────────────┐    │
│  │    Task Queue              │    │
│  │  • Task 1 ──────► Worker 1 │    │
│  │  • Task 2 ──────► Worker 2 │    │
│  │  • Task 3 ──────► Worker 3 │    │
│  │  • Task 4 ──────► Worker 4 │    │
│  └────────────────────────────┘    │
└─────────────────────────────────────┘

Performance Benefits

Workers	Speed Increase	Use Case
1 Worker	1x Baseline	Development
4 Workers	~3.8x Faster	Small cluster
10 Workers	~9.5x Faster	Medium cluster
50+ Workers	Linear Scaling	Cloud deployment

Formula: Speedup ≈ N / (1 + (N-1)*f) where f is parallelizable fraction

Features

• ✅ Dynamic worker joining/leaving
• ✅ Automatic task redistribution
• ✅ Load balancing
• ✅ Fault tolerance
• ✅ Minimal coordination overhead

---

Best Practices

1. Branch Naming: feature/, fix/, docs/, refactor/
2. Commit Messages: Use conventional commits
3. Code Reviews: All PRs require review
4. Testing: Distributed testing across multiple machines

---

🔮 Future Improvements

Planned Features

Feature	Status	Expected
Automatic Worker Discovery	Planned	2026
Advanced Load Balancing	Planned	2026
Fault Tolerance & Health Monitoring	Research	2026
Real-time Dashboard	Planned	2026
Docker/Kubernetes Deployment	Planned	2026
GPU Acceleration Support	Research	2026

Research Areas

• Machine Learning for optimal task distribution
• Blockchain for worker verification and trust
• Edge Computing integration for IoT devices
• Quantum Computing readiness

---

📞 Contact

Project Maintainer

👨‍💻 Zakaria HANANI

Platform	Link
GitHub	github.com/ZakariaHanani
Email	[email protected]
LinkedIn	linkedin.com/in/zakaria-hanani

👨‍💻 Mohamed OUIJJANE

Platform	Link
GitHub	github.com/MohamedOuijjane
Email	[email protected]
LinkedIn	[linkedin.com/in/mohamedouijjane

👨‍💻 Ayoub Karkouri

Platform	Link
GitHub	github.com/MohamedOuijjane
Email	[email protected]
LinkedIn	linkedin.com/in/ayoubkarkouri

👨‍💻 Ahmed LAHMAINE

Platform	Link
GitHub	github.com/ahmed-la14
Email	[email protected]

👨‍💻 Ali HALLA

Platform	Link
GitHub	github.com/Alihalla
Email	[email protected]

👨‍💻 Hmad AIT LAHMOUSS

Platform	Link
GitHub	github.com/hmad-ait-lahmous
Email	[email protected]

👨‍💻 Mohamed OUADRA

Platform	Link
GitHub	github.com/MohamedOuadra
Email	[email protected]

Contribution

We welcome contributions! Please:

1. Fork the repository
2. Create a feature branch
3. Submit a Pull Request
4. Join our discussions

---

---

🎯 Quick Start Summary

# 1. Clone & Build
git clone https://github.com/ZakariaHanani/distributed-cpu-montecarlo-traffic && cd repo
mvn clean install

# 2. Start Services (in separate terminals)
./scripts/run-registry.sh
./scripts/run-master.sh
./scripts/run-worker.sh  # Repeat for each worker

# 3. Launch Interface
cd frontend && npm run dev

# 4. Open browser: http://localhost:3000

---

📜 License

© 2025 – Distributed CPU Monte Carlo Traffic
Grid Computing • Java RMI • High-Performance Computing

---

🌟 Star us on GitHub if you find this project useful!

"Harnessing distributed power for complex simulations"

---

Tags: grid-computing monte-carlo java-rmi distributed-systems nextjs high-performance parallel-computing simulation

---

_{Built with ❤️ by the distributed computing community}