Python GIL Comparison

🚀 Exploring Python Without the GIL! 🚀

This repository benchmarks the performance of Python 3.13.1 with and without the Global Interpreter Lock (GIL) using real-world workloads.

📖 Read the full analysis and benchmark results in my blog post:
👉 Exploring Python 3.13: Hands-on with the GIL Disablement

Create the Docker Images

Build the Docker Image with GIL Enabled

docker build --no-cache --file python-3.13.1-bookworm-gil.Dockerfile --tag python:3.13.1-bookworm-gil .

Build the Docker Image with GIL Disabled

docker build --no-cache --file python-3.13.1-bookworm-nogil.Dockerfile --tag python:3.13.1-bookworm-nogil .

Check the GIL Status

This section verifies whether the Global Interpreter Lock (GIL) is enabled or disabled in the Python environment. Running gil_status.py will confirm if the GIL is active, ensuring the correct interpreter settings before executing performance benchmarks.

GIL-Enabled Image

Run gil_status.py to verify that the GIL is active:

docker container run --rm -it -v ${pwd}/src:/app -w /app python:3.13.1-bookworm-gil python gil_status.py

GIL-Disabled Image

Run gil_status.py to verify that the GIL is disabled:

docker container run --rm -it -v ${pwd}/src:/app -w /app python:3.13.1-bookworm-nogil python gil_status.py

Performance Comparison With and Without GIL

Prime Number Calculation

This benchmark tests a CPU-intensive mathematical computation by counting prime numbers within a given range. The test is executed in both single-threaded and multi-threaded modes to measure the impact of the GIL on parallel performance.

GIL Enabled

Run prime_counter.py to measure the performance of prime number calculations in Python 3.13.1 with the GIL:

docker container run --rm -it -v ${pwd}/src:/app -w /app python:3.13.1-bookworm-gil python prime_counter.py

GIL Disabled

Run prime_counter.py to measure the performance of prime number calculations in Python 3.13.1 with the GIL disabled:

docker container run --rm -it -v ${pwd}/src:/app -w /app python:3.13.1-bookworm-nogil python prime_counter.py

Loan Risk Scoring Benchmark

This benchmark simulates a real-world financial workload by analyzing loan applications and calculating a risk score for each applicant. The dataset is processed in both single-threaded and multi-threaded modes to compare performance with and without the GIL.

GIL Enabled

Run loan_risk_scoring_benchmark.py to measure the performance of risk scoring in Python 3.13.1 with the GIL:

docker container run --rm -it -v ${pwd}/src:/app -v ${pwd}/data:/data -w /app python:3.13.1-bookworm-gil python loan_risk_scoring_benchmark.py

GIL Disabled

Run loan_risk_scoring_benchmark.py to measure the performance of risk scoring in Python 3.13.1 with the GIL disabled:

docker container run --rm -it -v ${pwd}/src:/app -v ${pwd}/data:/data -w /app -e PYTHON_GIL=0 python:3.13.1-bookworm-nogil python loan_risk_scoring_benchmark.py