2.4. GPGPU ServerΒΆ

  1. After you log in the server, please run the following command to install the SDK in your home directory:

    $ cp_gpusdk
    

    This will result in a folder being created in ~/NVIDIA_GPU_Computing_SDK,

  2. Compiling the examples:

    $ cd ~/NVIDIA_GPU_Computing_SDK/C
    $ make
    
  3. Run the deviceQuery demo to see the GPGPU device(s).

    $ cd ~/NVIDIA_GPU_Computing_SDK/C/bin/linux/release/
    $ ./deviceQuery          
    [deviceQuery] starting...
    ./deviceQuery Starting...
    
    CUDA Device Query (Runtime API) version (CUDART static linking)
    
    Found 4 CUDA Capable device(s)
    ......
    
  4. See http://developer.download.nvidia.com/compute/cuda/3_2/docs/Getting_Started_Linux.pdf for additional details.