cudnn: Library of deep learning/neural network primitives for GPUs

Contents

  1. Overview of package
  2. Overview of package
    1. General usage
  3. Availability of package by cluster

Overview of package

General information about package
Package: cudnn
Description: Library of deep learning/neural network primitives for GPUs
For more information: https://developer.nvidia.com/cuda-zone
Categories:
License: Free2Use (NVIDIA Software License)

General usage information

cuDNN is a library of primitives supporting deep learning and neural networks using NVIDIA GPUs.

The following environmental variables have been defined:

  • \$CUDNN_ROOT has been set to the root of the cudnn installation
  • \$CUDNN_LIBDIR points to the directory containing the libraries
  • \$CUDNN_INCDIR points to the directory containing the header files

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

  • -I\$CUDNN_INCDIR
and the following flags to your link command (e.g. LDFLAGS in your Makefile):
  • -L\$CUDNN_LIBDIR -Wl,-rpath,\$CUDNN_LIBDIR

Available versions of the package cudnn, by cluster

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

Available versions of cudnn on the Deepthought2 cluster (RHEL8)

Available versions of cudnn on the Deepthought2 cluster (RHEL8)
Version Module tags CPU(s) optimized for GPU ready?
7.6.5.32-10.2-linux-x64 cudnn/7.6.5.32-10.2-linux-x64 ivybridge, x86_64, zen Y

Available versions of cudnn on the Juggernaut cluster

Available versions of cudnn on the Juggernaut cluster
Version Module tags CPU(s) optimized for GPU ready?
7.6.5.32-10.2 cudnn/7.6.5.32-10.2 skylake_avx512, x86_64, zen, zen2 Y
8.2.0.53-11.3 cudnn/8.2.0.53-11.3 x86_64 Y

Available versions of cudnn on the Deepthought2 cluster (RHEL6) [DEPRECATED]

Available versions of cudnn on the Deepthought2 cluster (RHEL6) [DEPRECATED]
Version Module tags CPU(s) optimized for GPU ready?
7.2.1
  • 7.2.1
  • (a.k.a cuDNN/v7.2.1)
x86_64 Y
5.1
  • 5.1
  • (a.k.a cuDNN/v5.1)
x86_64 Y
4.0
  • 4.0
  • (a.k.a cuDNN/v4.0)
x86_64 Y
v3.0 cuDNN/v3.0 x86_64 Y
v2 cuDNN/v2 x86_64 Y
v1 cuDNN/v1 x86_64 Y