Last Thursday, on 1 September, I held a talk at the GridKa Summer School 2016. Under the general topic of Data Science on Modern Architectures I introduced 30 Minutes programming on GPUs.

This was my first general introduction talk on the GPU platform, so I started from scratch (and tried only to be influenced as little as possible). Additionally, I made the slides completely in LaTeX Beamer. Using a lot of TikZ adds an additional layer of cumbersomeness on-top of the usual LaTeX cumbersome experience. 1

Anyway: I think the talk turned out quite well. I got some interested follow-up questions afterwards, especially in the coffee break. 2 The work was well worth it!

An embed of the slides follows. This is the handout version. You can also download the actual set I presented, which has more overlays to guide my story.

Your browser is unable to display the PDF. Instead, please download it.
  1. LaTeX is a constant struggle between »Why did I do this?« and »Wow, cool, that was worth it«. LaTeX, the alternating experience™. 

  2. Well, plus the usual discussion of NVIDIA vs. AMD.