FL Simulator

A Federated Learning simulation platform using the Flower framework with Docker-based deployment and real-time monitoring.

Last Updated: 2025-12-02 Version: 1.0.0

Features

• Docker-based FL Deployment: Containerized server and clients for easy deployment
• Real-time Monitoring: InfluxDB + Grafana stack with 5 pre-configured dashboards
• Battery SOH Use Case: Complete implementation for Battery State of Health prediction
• Scalable Architecture: Support for 1-4+ federated clients
• Automated Scripts: One-command training start/stop

Quick Start

# Clone and navigate to the project
cd fl-simulator/Battery_SOH

# Start federated learning with 4 clients
./start-fl-training.sh 4

# Access monitoring dashboards
open http://localhost:3005  # Grafana (admin/flsimulator2025)

Project Structure

fl-simulator/
├── README.md                      # This file
├── fl-simulator/
│   └── Battery_SOH/              # Battery SOH FL implementation
│       ├── README.md             # Project documentation
│       ├── server/               # FL Server
│       ├── client1-4/            # FL Clients (4 instances)
│       ├── monitoring/           # Grafana + InfluxDB stack
│       ├── start-fl-training.sh  # Training launcher
│       └── stop-fl-training.sh   # Stop all containers
└── specs/                        # Specifications

Architecture

┌─────────────────────────────────────────────────────────────┐
│                    FL SIMULATOR                             │
├─────────────────────────────────────────────────────────────┤
│                                                             │
│  ┌─────────────┐     ┌─────────────┐     ┌─────────────┐   │
│  │   Client 1  │     │   Client 2  │     │   Client 3  │   │
│  │  (Docker)   │     │  (Docker)   │     │  (Docker)   │   │
│  └──────┬──────┘     └──────┬──────┘     └──────┬──────┘   │
│         │                   │                   │           │
│         └───────────────────┼───────────────────┘           │
│                             │                               │
│                     ┌───────▼───────┐                       │
│                     │  FL Server    │                       │
│                     │  (Docker)     │                       │
│                     └───────┬───────┘                       │
│                             │                               │
│                     ┌───────▼───────┐                       │
│                     │   InfluxDB    │◄──── Metrics Push     │
│                     └───────┬───────┘                       │
│                             │                               │
│                     ┌───────▼───────┐                       │
│                     │   Grafana     │◄──── Visualization    │
│                     └───────────────┘                       │
│                                                             │
└─────────────────────────────────────────────────────────────┘

Monitoring Dashboards

Dashboard	Description
FL Global Training Progress	Real-time training overview with R², loss, and round progress
FL Client Comparison	Compare performance metrics across all clients
FL Round Analysis	Detailed round-by-round convergence analysis
FL Historical Analysis	Long-term trends and historical data
FL Experiment Comparison	Compare different training experiments

Access all dashboards at: http://localhost:3005

Use Cases

Battery SOH Prediction

Located in fl-simulator/Battery_SOH/:

• Federated learning for battery health prediction
• 4 clients with distributed battery data
• scikit-learn based model training
• R² and MSE loss tracking

See Battery_SOH/README.md for details.

Requirements

• Docker Engine 20.10+
• Docker Compose v2+
• Python 3.10+ (for local development)
• 4GB+ RAM recommended

Configuration

Server Settings

Edit fl-simulator/Battery_SOH/server/pyproject.toml:

[tool.flwr.app.config]
num-server-rounds = 10      # Number of FL rounds
fraction-fit = 1.0          # Fraction of clients for training
fraction-evaluate = 1.0     # Fraction of clients for evaluation

Monitoring Settings

Edit fl-simulator/Battery_SOH/monitoring/docker-compose.yml for:

• Port mappings
• Resource limits
• Retention policies

Ports Used

Service	Port	Description
Grafana	3005	Monitoring dashboards
InfluxDB	8086	Time-series database
FL SuperLink	9092	Flower server communication

Documentation

• Battery SOH README - Main project documentation
• Monitoring README - Monitoring stack details

Contributing

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Submit a pull request

License

MIT License - See LICENSE file for details.

Changelog

2025-12-02

• Fixed all Grafana dashboards to use InfluxDB Flux queries
• Migrated from Prometheus to InfluxDB-only architecture
• Added 5 comprehensive monitoring dashboards
• Updated documentation across all READMEs

2025-11-29

• Initial project setup
• Docker-based FL deployment
• Basic monitoring with InfluxDB + Grafana