Main Menu

cuda: parallel computing platform/programming model for NVIDIA GPUs

Contents

  1. Overview of package
    1. General usage
  2. Availability of package by cluster
  3. Driver versions
  4. Notes on using Cuda
  5. Notes on using OpenCL

Overview of package

General information about package
Package: cuda
Description: parallel computing platform/programming model for NVIDIA GPUs
For more information: https://developer.nvidia.com/cuda-zone
Categories:
License: Unknown

General usage information

CUDA is a parallel computing platform and programming model designed by NVIDIA enabling one to harness the power of NVIDIA GPUs.

This module will add the nvcc and other CUDA commands to your path.

The following environmental variables have been defined:

You will probably wish to use these by adding the following flags to your compilation command (e.g. to CFLAGS in your Makefile):

and the following flags to your link command (e.g. LDFLAGS in your Makefile):

Available versions of the package cuda, by cluster

This section lists the available versions of the package cudaon the different clusters.

Available versions of cuda on the Zaratan cluster

Available versions of cuda on the Zaratan cluster
Version Module tags CPU(s) optimized for GPU ready?
10.2.89 cuda/10.2.89 zen Y
11.6.2 cuda/11.6.2 zen2 Y
12.3.0 cuda/12.3.0 icelake, x86_64, zen2 Y

NVIDIA Drivers and versions

The versions listed above are the versions of the CUDA libraries installed in the Research Software Library for the relevant cluster. The GPU-enabled nodes also have the relevant GPU drivers installed, which are differently versioned, but the drivers installed on the GPU enabled nodes support at least the specified versions of CUDA. You are free to install newer versions of the CUDA libraries for personal use in directories you have write access to (e.g. your home or scratch directories), up to the latest versions supported by the GPU drivers installed on the GPU-enabled nodes. You will not be able to change the version of the installed GPU drivers.

The GPU-enabled nodes on Zaratan currently are running:

Available versions of GPU drivers by cluster
Cluster Driver Version Latest CUDA Version supported by driver
Zaratan NVIDIA Driver 560.28.03 CUDA 12.6

Notes on using CUDA

nvcc can compile source files with the CUDA language extensions. It also contains CUDA implementations of BLAS and FFT libraries, as well as some data parallel primitives. See the Wikipedia article on CUDA for general information, or NVIDIA CUDA documentation for detailed instructions.

Notes on using OpenCL

OpenCL is a framework for writing programs to execute across CPUs and various other devices (including GPUs). See the Wikipedia article on OpenCL for more information.

WARNING
Deepthought2 RHEL6 Users: The OpenCL support in cuda/5.0.35 appears to be broken. Use cuda/6.5.14 or newer instead.

The nvcc compiler will support OpenCL as well. Your code should include the opencl.h header file, e.g. 

#include <CL/opencl.h>
and you should compile it with something like 
nvcc -lOpenCL mycode.c





Back to Top