CUDA

CUDA is NVIDIA parallel computing architecture that enables dramatic increases in computing performance by harnessing the power of the GPU (graphics processing unit). Currently CUDA 10.0 compatible NVIDIA drivers are installed (version 410.78). There are several versions of CUDA available,

To use CUDA 7.5.18 you need to add the following environment variables to your.bashrc file,

export PATH=/usr/local/cuda-7.5/bin:$PATH
export LD_LIBRARY_PATH=/usr/local/cuda-7.5/lib64:/usr/local/cuda-7.5/lib:$LD_LIBRARY_PATH

Or load the environment module for CUDA 7.5 by executing the following command,

module load cuda/7.5

CUDA documentation can be found in /usr/local/cuda-7.5/doc

CUDA Samples can be found in /usr/local/cuda-7.5/samples

To use CUDA 8.0.61 you need to add the following environment variables to your .bashrc file,

export PATH=/usr/local/cuda-8.0/bin:$PATH
export LD_LIBRARY_PATH=/usr/local/cuda-8.0/lib64:/usr/local/cuda-8.0/lib:/usr/local/cuda-8.0/extras/CUPTI/lib64:$LD_LIBRARY_PATH

Or load the environment module for CUDA 8.0, by executing the following command,

module load cuda/8.0

CUDA documentation can be found in /usr/local/cuda-8.0/doc

CUDA Samples can be found in /usr/local/cuda-8.0/samples

To use CUDA 9.0.176 you need to add the following environment variables to your .bashrc file or shell,

export PATH=/usr/local/cuda-9.0/bin:$PATH
export LD_LIBRARY_PATH=/usr/local/cuda-9.0/lib64:/usr/local/cuda-9.0/extras/CUPTI/lib64:$LD_LIBRARY_PATH

Or load the environment module for CUDA 9.0, by executing the following command,

module load cuda/9.0

CUDA documentation can be found in /usr/local/cuda-9.0/doc
CUDA Samples can be found in /usr/local/cuda-9.0/samples
To use CUDA 9.1.85 you need to add the following environment variables to your .bashrc file or shell,

export PATH=/usr/local/cuda-9.1/bin:$PATH
export LD_LIBRARY_PATH=/usr/local/cuda-9.1/lib64:/usr/local/cuda-9.1/extras/CUPTI/lib64:$LD_LIBRARY_PATH

Or load the environment module for CUDA 9.1, by executing the following command,

module load cuda/9.1

CUDA documentation can be found in /usr/local/cuda-9.1/doc
CUDA Samples can be found in /usr/local/cuda-9.1/samples
To use CUDA 9.2.148 you need to add the following environment variables to your .bashrc file or shell,

export PATH=/usr/local/cuda-9.2/bin:$PATH
export LD_LIBRARY_PATH=/usr/local/cuda-9.2/lib64:/usr/local/cuda-9.2/extras/CUPTI/lib64:$LD_LIBRARY_PATH

Or load the environment module for CUDA 9.2, by executing the following command,

module load cuda/9.2

CUDA documentation can be found in /usr/local/cuda-9.2/doc

CUDA Samples can be found in /usr/local/cuda-9.2/samples

To use CUDA 10.0.130 you need to add the following environment variables to your .bashrc file or shell,

export PATH=/usr/local/cuda-10.0/bin:$PATH
export LD_LIBRARY_PATH=/usr/local/cuda-10.0/lib64:/usr/local/cuda-10.0/extras/CUPTI/lib64:$LD_LIBRARY_PATH

Or load the environment module for CUDA 10.0, by executing the following command,

module load cuda/10.0

CUDA documentation can be found in /usr/local/cuda-10.0/doc
CUDA Samples can be found in /usr/local/cuda-10.0/samples

Running NVIDIA Nsight

NVIDIA Nsight is an IDE based on Eclipse for CPU and GPU code. To run Nsight select from the Applications menu → Software → NSight

NOTE: NSight will use CUDA 9.0 unless another CUDA environment variables are set.

你可能感兴趣的:(CUDA)