Python Web Benchmarks

Real-world performance benchmarking of Python web applications across different Python versions.

🎯 Project Overview

This project benchmarks Wagtail CMS (a popular Django-based CMS) across multiple Python versions to provide actionable insights for version migration decisions.

Supported Python Versions:

• Python 3.10, 3.11, 3.12, 3.13, 3.14
• Python 3.13t (free-threaded/nogil)
• Python 3.14t (free-threaded/nogil)
• Python 3.14-tailcall (experimental tailcall optimization)

📊 Latest Results

Python 3.11 shows ~18% average performance improvement over Python 3.10!

See BENCHMARK_SUMMARY.md for detailed results.

Quick Stats

• Python 3.11: Best overall performance (+18% vs 3.10)
• Python 3.12: Competitive with 3.11
• Memory Usage: Similar across all versions (~270-287 MB)
• Success Rate: 100% across all tests

🚀 Quick Start

Prerequisites

• Docker installed
• Python 3.x
• Apache Bench (ab)

Run Benchmarks

# Build Docker images for all Python versions
make build

# Run the benchmark suite
make run

# Generate summary report
python3 scripts/generate_summary.py

📁 Project Structure

python-web-benchmarks/
├── apps/
│   └── wagtail/              # Wagtail CMS configuration
│       ├── Dockerfile
│       ├── setup_realworld_data.py  # Real-world data generator
│       └── entrypoint.sh
├── benchmarks/
│   ├── runner.py             # Main orchestration
│   ├── metrics_collector.py  # Resource monitoring
│   └── load_generator.py     # HTTP load testing
├── config/
│   └── config.yaml          # Benchmark configuration
├── data/
│   ├── raw/                 # JSON results
│   └── BENCHMARK_SUMMARY.md # Summary report
├── scripts/
│   └── generate_summary.py  # Report generator
└── Makefile                 # Build and run commands

🧪 Test Scenarios

The benchmark includes real-world test scenarios:

1. Homepage Load - Full page with navigation, sections, and content
2. Admin Login - Authentication and admin interface
3. Page Browsing - Blog pages with 15+ articles

Each scenario is tested with:

• 1, 5, 10, and 20 concurrent users
• Multiple requests per user
• Full resource monitoring

📈 Real-World Data

The Wagtail instance includes:

• ✅ 15+ blog posts with realistic content
• ✅ Multiple pages (About, Services, Contact, etc.)
• ✅ Fully configured admin interface
• ✅ Real database with migrations
• ✅ Static file serving

🔧 Configuration

Edit config/config.yaml to:

• Add/remove Python versions
• Modify test scenarios
• Adjust concurrent user levels
• Change test duration

📊 Results Format

Results are saved as JSON in data/raw/ with metrics including:

• Requests per second (RPS)
• Response times
• Error rates
• CPU usage
• Memory consumption

🎨 Available Commands

make help      # Show available commands
make install   # Install Python dependencies
make build     # Build Docker images
make run       # Run benchmark suite
make clean     # Clean up containers and images

🏆 Key Findings

Based on the latest benchmark run:

1. Performance: Python 3.11 provides the best overall performance
2. Stability: 100% success rate across all Python versions
3. Resources: Memory usage is consistent across versions
4. Recommendation: Upgrade to Python 3.11 or 3.12 for production workloads

📝 License

MIT

🤝 Contributing

Contributions welcome! Areas for improvement:

• Add more web applications (Airflow, Sentry, etc.)
• Add visualization dashboards
• Implement CI/CD pipelines
• Test additional Python experimental features