World Congress 2024 • Aug 20, 2024 • Session details

Accelerating Python on GPUs

Paul Graham

Think maximizing GPU power requires low-level CUDA syntax? Think again. Scale your Python computations using top-down frameworks, drop-in acceleration libraries, and JIT compilation for massive performance gains.

Pause
Mute Enter Fullscreen
#1 about 5 min

Introduction to hardware parallelism and GPU computing

The necessity of parallel processing hardware to overcome limitations in single-core CPU architectures.

#2 about 2 min

Navigating the CUDA ecosystem and abstraction layers

Evaluating different entry points for accelerated compute from high-level applications down to custom compilers.

#3 about 6 min

Leveraging specialized AI frameworks and accelerated libraries

Replacing standard data science and simulation packages with optimized alternatives to gain order-of-magnitude execution speeds.

#4 about 9 min

Writing direct GPU code using Python extensions

Controlling array computation and explicit data movement through drop-in replacements and dynamic compilation utilities.

#5 about 2 min

Profiling and debugging parallel Python computing applications

Utilizing ecosystem tools like Nsight to identify system bottlenecks and prevent memory race conditions.

#6 about 2 min

Accessing software containers and developer training platforms

Retrieving containerized deep learning packages and hardware compiler stacks to simplify accelerated deployments.

Matching moments

4:37 min

Simplifying parallel programming with the CUDA ecosystem

Paul Graham Paul Graham · LIVE

3:32 min

Evolution of general purpose GPU computing and Python

Paul Graham Paul Graham · World Congress 2025

1:30 min

Exploring the Python-enabled GPU programming software stack

Paul Graham Paul Graham · World Congress 2026 Europe

1:49 min

Accelerating script execution with CuPy and Numba kernels

Paul Graham Paul Graham · World Congress 2026 Europe

1:37 min

Accelerating compute with focused developer tools

Julia Koch Julia Koch +1 · World Congress 2026 Europe

2:15 min

Introduction to CUDA and general-purpose GPU computing

Paul Graham Paul Graham · World Congress 2026 Europe