ÎÛÎÛ²ÝÝ®ÊÓƵ

Event

Introduction à la programmation GPU avec CUDA [CQ - UdeM]

Tuesday, March 17, 2020 09:00to16:00
Pavillon Jean-Coutu, S1-125, Université de Montréal, 2940 Chemin de Polytechnique, Montreal, QC, H3T 1J4, CA
Price: 
$10 - $350

This workshop will be given in French by Mr. Nikolai Sergueev, HPC analyst at Calcul Québec - l'Université de Montréal.

Summary

CUDA, which stands for Compute Unified Device Architecture, is the main programming language used to harness the power of GPU (Graphic Processing Units) in high performance computing. It is supported on every NVidia GPU accelerators. In this tutorial, you will learn how to use CUDA with the programming language C, to write simple algorithms on GPUs.

Registration

Academic participant : 10$

Non-academic participant : $350 

Prerequisite
Knowledge of the C/C++ programming language is required. Attendees should bring their personal laptop computer to participate in the hands-on content.

³¢±ð²õ²õ´Ç²ÔÌý±è±ô²¹²Ô

Architecture of a GPU

Workflow of a CUDA program

Core principles of using GPU efficiently

Writing and compiling a minimal CUDA code and compiling CUDA examples

Managing GPU memory and understanding the various types of GPU memory

Using threads and blocks to write parallel algorithms

Checking errors

Overview of what’s outside CUDA (OpenACC, Libraries, OpenCL, Applications)

Optional advanced topics (multiple GPUs, unified memory, Nvidia Visual Profiler, streams)

How can I contact the organizer if I have any questions?

You may write to formation [at] calculquebec.ca

Back to top