Info for Students

NVIDIA: Basics of accelerated computing with CUDA C / C ++

Publish date: 2021/02/11 | Expire date: 2021/02/18

News is expired!

Duration: 10:00 - 17:30

Where: MS Teams (link will be sent to participants on the day of the workshop)

In this workshop doc. dr. Domen Verber, ambassador for NVIDIA, will present the fundamental tools and techniques for accelerating C/C++ applications to run on massively parallel GPUs with CUDA ® . You’ll learn how to write code, configure code parallelization with CUDA, optimize memory migration between the CPU and GPU accelerator, and implement the workflow that you’ve learned on a new task—accelerating a fully functional, but CPU-only, particle simulator for observable massive performance gains.

You will learn:

  • how to write code,
  • configure the parallelization code with CUDA, and
  • optimize memory migrations between CPU and GPU accelerator.

Finally, you will implement the learned workflow on the following task: accelerating a fully functional (CPE) particle simulator for the observed increased performance.

Upon successful completion of the assessment, participants will receive an NVIDIA DLI certificate to recognize their subject matter competency and support professional career growth.

Goals

After completing the workshop, you will understand the basic tools and techniques for C / C ++ applications that accelerate GPU with CUDA and can:

  • write the code to be executed by the GPU accelerator,
  • highlight and express the parallelism of data and parallelism at the level of instructions in C / C ++ applications using CUDA,
  • use CUDA-managed memory and optimize memory migration using asynchronous prefetching,
  • use the command line and visual profiles to direct your work,
  • use concurrent flows for instruction-level parallelism, and
  • write GPU-accelerated CUDA C/C ++ applications, or redesign existing CPU-accelerated applications using a profile-based approach.

Preconditions

Basic C/C++ competency, including familiarity with variable types, loops, conditional statements, functions, and array manipulations. No previous knowledge of CUDA programming is assumed.

Program

10:00- 10:30

Presentation of student organizations (IEEE Student Branch Maribor, IEEE Student Branch Ljubljana, IEEE Women in Engineering Slovenia, ACM Student Chapter Maribor)

10:15 – 10:30

Introduction

10:30 - 12.30

Accelerating applications with CUDA C / C ++

12:30 – 13:00

Break

13:00 – 15:00

Memory management of accelerated applications with CUDA C / C ++

15:00 – 15:15

Break

15:15 – 17:15

Asynchronous streaming and visual profiling for accelerated applications with CUDA C / C ++

17:15 – 17:30

Closing of the workshop

Registration for the workshop

Number of places: 30

Price: The workshop is originally intended for students of UM, UL and UPR, but university employees and IEEE members are also welcome.

Registration deadline: February 16, 2021 at 3 p.m.

Registration form: Microsoft Forms

The workshop is organized in cooperation with the following organizations: