CUDA Programming

CUDA (or Compute Unified Device Architecture) is a parallel computing platform and application programming interface (API) that allows software to use certain types of graphics processing unit (GPU) for general purpose processing Wikipedia

Prerequisite

If you want to compile and run the code in your machine, the following conditions are required:

• Machine with Nvidia GPU (CUDA Supported)
• CUDA SDK -> Download

Installation CUDA SDK (openSUSE)

• Open the Download page and download installer according to your OS
• Add Repository, refresh, and install cuda

$ sudo zypper addrepo https://developer.download.nvidia.com/compute/cuda/repos/opensuse15/x86_64/cuda-opensuse15.repo
$ sudo zypper refresh
$ sudo zypper install -y cuda

• Add cuda binary location to PATH variable

$ export PATH=/user/local/cuda/bin:$PATH

• Verify Installation

$ nvcc --version

Resource

If you want to learn more about CUDA you can refer to this resource

• Cuda Sample Code