Publication date : 03/26/2024

Course : C++, parallel programming with CUDA in Visual Studio with NVIDIA cards

use the GPU to improve performance

Practical course - 2d - 14h00 - Ref. CDU
Price : 1280 € E.T.

C++, parallel programming with CUDA in Visual Studio with NVIDIA cards

use the GPU to improve performance



In this training course, you will discover, evaluate and manipulate the CUDA SDK from NVIDIA, a leader in the use of the GPU to improve data parallelism performance. You'll acquire all the knowledge you need to implement CUDA.


INTER
IN-HOUSE
CUSTOM

Practical course in person or remote class
Disponible en anglais, à la demande

Ref. CDU
  2d - 14h00
1280 € E.T.




In this training course, you will discover, evaluate and manipulate the CUDA SDK from NVIDIA, a leader in the use of the GPU to improve data parallelism performance. You'll acquire all the knowledge you need to implement CUDA.


Teaching objectives
At the end of the training, the participant will be able to:
Understand the benefits of using the GPU as an independent computing resource
Using the GPU with CUDA in the Visual Studio environment
Linking C++11 threads to GPU usage
Verify interest as part of a complete project

Intended audience
C/C++ application designers and developers, software architects.

Prerequisites
Good knowledge of C/C++ and threads, experience required. Basic knowledge of C++11.

Practical details
Hands-on work
Development of a test application, evaluation of the different solutions proposed and comparison with equivalent CPU processing.
Teaching methods
Theoretical presentations followed by code review and implementation on a test application.

Course schedule

1
Introduction

  • GPU usage compared to CPU.
  • The CUDA SDK from NVIDIA.
  • Alternatives and complements to CUDA.
Demonstration
Presentation of the test application and evaluation of CPU results.

2
CUDA installation

  • Installation of the specific CUDA driver and SDK.
  • Installation of NSIGHT, the CUDA-specific environment in Visual Studio.
  • Exploring application examples.
  • Recovery of installed graphics card capacities.
Hands-on work
CUDA installation, project creation and validation.

3
Basic implementation

  • The fundamentals of kernel function execution.
  • Creating a kernel function.
  • Calling a kernel function.
  • Memory transfers between host and GPU.
  • Asynchronous execution of a GPU code sequence.
  • Debugging code executed on the GPU.
Hands-on work
Add a code sequence to be executed on the GPU to the test application, and compare the results with the existing C++11 code. Use of the NSIGHT debugger.

4
Using CUDA's different memory options

  • Shared memory within a thread block, different options.
  • Optimization between memory dedicated to data and the size of the code to be executed.
  • Mapped allocations between host memory and graphics card memory.
  • The use of portable memory between the host and several graphics cards.
Hands-on work
Manipulate the different options in the test application. Find the best solution for a given case.

5
Other CUDA applications

  • The use of Streams, parallel execution on different graphics cards.
  • Using CUDA in C++ with Thrust.
  • CUDA alternatives or complements such as C++ AMP, OpenCL, OpenAPP.
Case study
Exploration of complementary and alternative solutions, comparison using the test application.

6
Conclusion

  • The scope for using the GPU as an alternative to the CPU.
  • Best practices.


Customer reviews
4,6 / 5
Customer reviews are based on end-of-course evaluations. The score is calculated from all evaluations within the past year. Only reviews with a textual comment are displayed.
MEHDI R.
25/09/25
4 / 5

PCs sometimes froze
EDOUARD S.
25/09/25
5 / 5

It's a very dense course. The ratio of theory to practical work is perfect.
COLAS J.
25/09/25
5 / 5

Clear, detailed training. Quite comprehensive for an introduction.



Dates and locations
Select your location or opt for the remote class then choose your date.
Remote class

Dernières places
Date garantie en présentiel ou à distance
Session garantie

REMOTE CLASS
2026 : 30 Mar., 11 June, 26 Oct.

PARIS LA DÉFENSE
2026 : 30 Mar., 11 June, 26 Oct.