CUDA Frequently Asked Questions


  1. What is CUDA?

  2. CUDA stands for Compute Unified Device Architecture. CUDA is a parallel computing architecure and C based programming language for general purpose computing on NVIDIA GPU's

  3. Which department machines support CUDA?

  4. A list of machines with higher-end cuda cards can be found on the CS filesystem at ~info/cuda_machines.

  5. How do I configure my shell environment for CUDA development?

  6. The NVIDIA graphics driver and CUDA compilier are already installed on machines that support CUDA. However, one must set some environment variables in order to run and write CUDA enabled programs.

    If you use tcsh, add the following lines to the bottom of your .cshrc file:

            # add cuda tools to command path
            set path = ( /usr/local/cuda/latest/bin $path )
            setenv MANPATH /usr/local/cuda/latest/man:${MANPATH}
    
            # add cuda libraries to library path
            if ($?LD_LIBRARY_PATH) then
              setenv LD_LIBRARY_PATH /usr/local/cuda/latest/lib64:${LD_LIBRARY_PATH}
            else
              setenv LD_LIBRARY_PATH /usr/local/cuda/latest/lib64
            endif
          

    If you use bash, add the following lines to the bottom of your .bashrc file:

            # add cuda tools to command path
            export PATH=/usr/local/cuda/latest/bin:${PATH}
            export MANPATH=/usr/local/cuda/latest/man:${MANPATH}
    
            # add cuda libraries to library path
            if [[ "${LD_LIBRARY_PATH}" != "" ]]
            then
              export LD_LIBRARY_PATH=/usr/local/cuda/latest/lib64:${LD_LIBRARY_PATH}
            else
              export LD_LIBRARY_PATH=/usr/local/cuda/latest/lib64
            fi
          

    Log out and back in for the changes to take effect.

  7. Where Can I find more information on CUDA?

  8. Go go The CUDA Zone