Loading…
Attending this event?
View analytic
Wednesday, September 26 • 16:45 - 17:45
Accelerating Applications on a GPU with CUDA C++

Sign up or log in to save this to your schedule and see who's attending!

CUDA is a parallel computing platform for a generic computing on Nvidia GPUs. We will focus on CUDA C/C++ programming, covering in detail various types of GPU memory (Shared/Global/Texture/ConstShared/Pinned), optimization of the work distribution between CUDA “threads”, precompiled CUDA code VS Runtime Compilation.
Often the CUDA code is written in C style, missing the opportunity to use safer and more readable C++ constructs. In the spirit of the conference a special focus will be given to the use of modern C++ features in the CUDA applications inside the device (“kernel”) code, host code, and more importantly – which of the C++ paradigms can cross the bridge between the two worlds.

Speakers
avatar for Michael Gopshtein

Michael Gopshtein

Framework TL, Eyeway Vision
Michael is a software engineer with more that 15 years of C++ experience. He worked on various performance-bound projects, like load generation on servers, network sniffing and packet analysis. In the recent years Michael is focused on Augmented Reality challenges in Eyeway Vision... Read More →


Wednesday September 26, 2018 16:45 - 17:45
F
  • Level Beginner, Intermediate
  • Tags CUDA, GPUs

Attendees (29)