README.md

Module 098: Performance Optimization & Profiling

• Phase: 10. Concurrency & Internals
• Duration: 2 hours

Learning Objectives

• Profile Python code with cProfile and profile
• Analyze profiling results with pstats
• Benchmark small code snippets with timeit
• Identify and fix performance bottlenecks
• Understand algorithmic complexity (Big O)
• Use slots for memory optimization
• Compare list vs array vs set performance

Topics Covered

1. Profiling with cProfile and profile
2. pstats for analysis
3. timeit module for micro-benchmarks
4. Identifying bottlenecks
5. Algorithmic optimization (Big O)
6. Using slots
7. List vs array vs set performance
8. C extensions briefly (Cython, cffi)
9. PyPy as alternative interpreter

Prerequisites

Modules 000-097.

Key Concepts

import cProfile
import pstats
import timeit
from typing import List

# Profiling
def slow_function(n: int) -> int:
    total: int = 0
    for i in range(n):
        total += i * i
    return total

cProfile.run('slow_function(1000000)', 'profile_stats')
p = pstats.Stats('profile_stats')
p.sort_stats('cumtime').print_stats(10)

# Micro-benchmarks
time: float = timeit.timeit(
    'sum(range(1000))',
    number=10000
)
print(f"Average: {time / 10000:.6f}s")

Resources

• Python profiling documentation
• cProfile docs: https://docs.python.org/3/library/profile.html
• timeit docs: https://docs.python.org/3/library/timeit.html
• Cython documentation: https://cython.readthedocs.io
• PyPy: https://pypy.org

Next Module

Module 099: Capstone Project: Full-Stack Application