The art of wrangling your GPU Python environments
summary
Benchmarks claim pandas work that takes ~59 seconds drops to 742 milliseconds on the GPU, and the talk shows a concrete utility called RAPIDS Doctor that runs command‑line checks for GPU presence, driver and CUDA versions, OS libraries, and package mismatches. The tool reports specific failures, gives exact upgrade or install steps, supports plugin checks per library, and demos a broken vs healthy environment. Roadmap includes Docker/Kubernetes and cloud integrations; the build is functional as a diagnostic CLI.
AI-written summary · about
chapters (8)
- 0:00Welcome
- 1:34Rapids project overview
- 6:32GPU Python setup challenges
- 8:18Python GPU ecosystem improvements
- 13:58Rapid Doctor introduction
- 16:11Rapid Doctor diagnostics
- 18:17Rapid Doctor future plans
- 19:36Q&A
related talks
29mA Beginners Guide to GPUs for Pythonistas
01h 12mGPU-Accelerated Data Analytics in Python
02h 28mGPU Accelerated Python
41mLuxuries, necessities, and the challenges that remain
01h 22mHigh-Performance Data Science at Scale with RAPIDS, Dask, and GPUs
41mAccelerating Data Science with RAPIDS
30mCUDA in Your Python: Effective Parallel Programming on the GPU
55mAccelerated Python: The Community and Ecosystem