Harness the Power of the GPU with PyCUDA (and Friends)
summary
Demonstrates how to harness GPU computing from Python using PyCUDA and related tools. Covers using CUDA kernels, managing GPU memory, and integrating PyCUDA with Python workflows to accelerate numerical and data-parallel tasks.
AI-written summary · about
chapters (18)
- 0:00PyCUDA for GPU computing
- 2:03GPUs vs CPUs
- 4:35GPU basics and performance
- 6:19CUDA vs OpenCL
- 8:28Why CUDA over OpenCL
- 10:19CUDA programming model
- 13:31PyCUDA programming
- 14:58Vector addition with PyCUDA
- 23:20Memory bandwidth limits
- 25:05Optimizing CUDA kernels
- 27:13Parallel reduction case study
- 28:59Numba for GPU computing
- 31:12Numba limits and setup
- 33:25Numba syntax constraints
- 37:17Theano for GPU math
- 40:11Theano GPU features
- 42:47GPU tools summary
- 43:32Q&A
related talks
30mCUDA in Your Python: Effective Parallel Programming on the GPU
35mPython at the Speed of Light: A Complete Tour of the CUDA Ecosystem
02h 28mGPU Accelerated Python
35mCUDA in Python- A New Era for GPU Acceleration
30mCuPy: A NumPy-compatible Library for GPU
48mGPU Programming in Pure Python
34mAccelerating GPU Algorithms in pure Python
55mAccelerated Python: The Community and Ecosystem