Hands-On CUDA 101

This meetup was held in Santa Clara on March 20, 2018.

GPUs are getting more and more used in the AI/ML field for computations, but creating GPU enabled applications differs from writing CPU programs. This workshop will give a high-level overview of GPU hardware, point out why is it a perfect match for certain ML algorithms (and why might it be a bad idea in other cases) and lay down the CUDA programming philosophy.

This is an introductory workshop where we’ll cover all the basics from how a CUDA C/C++ program is structured, what are its main components, how to write and compile simple kernels and how will they be run on the GPU.

We’ll also, time allowing, go through most common GPU libraries such as Thrust or cuBLAS.

This will be a hands-on laboratory so please bring a laptop with you! You are welcome to bring a machine with a Nvidia GPU, CUDA and Jupyter Notebook installed or you can use one of the cloud instances we’ll provide via https://qwiklabs.com/

Speaker’s Bio:

Mateusz is a software developer who loves all things distributed, machine learning and hates buzzwords. His favorite hobby data juggling.

He obtained his M.Sc. in Computer Science from AGH UST in Krakow, Poland, during which he did an exchange at L’ECE Paris in France and worked on distributed flight booking systems. After graduation, he moved to Tokyo to work as a researcher at Fujitsu Laboratories on machine learning and NLP projects, where he is still currently based.

In his spare time, he tries to be part of the IT community by organizing, attending and speaking at conferences and meetups.

Видео Hands-On CUDA 101 канала H2O.ai