Performance Comparison: C, C++, Python, and Java

This project compares the runtime performance of iterative computations in C, C++, Python, and Java. It demonstrates the use of Python's ctypes library for cross-language integration and highlights the performance differences between compiled and interpreted languages.

Table of Contents

1. Overview
2. Features
3. Requirements
4. Setup
5. How to Run
6. Results
7. Lessons Learned
8. License

Overview

This project explores runtime differences between C, C++, Python, and Java by:

• Writing shared libraries (.so files) in C and C++.
• Calling these libraries from Python using the ctypes module.
• Measuring and comparing runtimes across these languages.
• Integrating Java for an additional runtime comparison.

Features

• Cross-language integration using ctypes and subprocess.
• Performance benchmarking for iterative computations.
• Logarithmic runtime comparison plots.
• Demonstrates Java's runtime performance alongside C, C++, and Python.

Requirements

• GCC or Clang (for compiling C and C++ files)
• Python 3.x
• Java Development Kit (JDK)
• Python modules:
  • matplotlib

Setup

1. Clone the repository:

   git clone https://github.com/your-username/performance-comparison-c-python-cpp-java.git
   cd performance-comparison-c-python-cpp-java
   

2. Compile the C and C++ source files into shared libraries:

   gcc -shared -o compiled/loop.so -fPIC src/loop.c
   g++ -shared -o compiled/loop_cpp.so -fPIC src/loop.cpp
   

3. Compile the Java file:

   javac -d compiled src/Loop.java
   

4. Install Python dependencies:

   pip install matplotlib
   

How to Run

Compare Runtimes

1. Run the compare_runtime.py script to measure runtimes:

   `python scripts/compare_runtime.py`
   

Generate Runtime Plot

2. Run the compare_runtime_plot.py script to visualize runtimes:

   `python scripts/compare_runtime.py`
   

Example Output

• Text Output: The compare_runtime.py script prints the runtime for each language along with the final result.

  Running loop in C...
  C runtime: 0.000040 seconds, Result: 10001.000010
  Running loop in Python...
  Python runtime: 0.000397 seconds, Result: 10001.000010
  Running loop in C++...
  C++ runtime: 0.000015 seconds, Result: 10001.000010
  Running loop in Java...
  Java runtime: 0.000045 seconds, Result: 10001.000010

• Plot Output: The compare_runtime_plot.py script generates a runtime_comparison.png file showing the runtime comparison on a logarithmic scale.

Results

• C and C++: Near-identical runtimes due to similar compilation processes.
• Java: Competitive runtime due to Just-In-Time (JIT) compilation but slightly slower than C/C++.
• Python: Significantly slower, demonstrating the overhead of interpreted languages.

Lessons Leanred

• Cross-Language Integration: Combining Python with C/C++/Java leverages performance and flexibility.
• Performance Tradeoffs: Compiled languages (C, C++, Java) excel in performance, while Python offers ease of development.
• Effective Use of Tools: ctypes and subprocess enable seamless integration across languages.

License

This project is licensed under the MIT License.