Explore how to overcome the limitations of Python's Global Interpreter Lock (GIL) using asyncio in this 44-minute PyCon US talk by Łukasz Langa. Discover alternative approaches to multiprocessing for utilizing multiple CPU cores efficiently, particularly when dealing with large data transfers. Examine two practical examples of data processing using Python 3.11, demonstrating how asyncio combined with shared memory can enhance performance. Learn to leverage Python's built-in components to construct scalable systems, with demonstrations on both macOS and Windows Subsystem for Linux. Focus on core language features without delving into data pipeline reimplementation or MLops best practices.
Overview
Syllabus
Talks - Łukasz Langa: Working Around the GIL with asyncio
Taught by
PyCon US