CuDNN library

Z MetaCentrum
Přejít na: navigace, hledání

Description

The NVIDIA CUDA® Deep Neural Network library (cuDNN) is a GPU-accelerated library of primitives for deep neural networks. cuDNN provides highly tuned implementations for standard routines such as forward and backward convolution, pooling, normalization, and activation layers. cuDNN is part of the NVIDIA Deep Learning SDK.

Category

Development tools and environments

Availability

Versions and modules:

cudnn-7.0                  (for CUDA 8.0 and later)
cudnn-6.0                  (for CUDA 7.5 and later)
cudnn-5.1                  (for CUDA 7.5 and later)
cudnn-5.0                  (for CUDA 7.5 and later)
cudnn-4.0                  (for CUDA 7.0 and later)

Notice: This is a licenced software. If you want to use it, you must confirm the licence form. But first you have to accept a licence at NVIDIA's site.
Notice 2: These are the standalone modules. Usually you need to use it with some of CUDA modules.

Supporting GPU clusters

CuDNN only works on GPUs with high enough computing capabilities. In this table, you can see information about individual GPU clusters and if their GPUs support CuDNN library:


GPU clusters in MetaCentrum
Cluster Nodes GPUs per node Compute Capability CuDNN gpu_cap=
doom.metacentrum.cz doom1.metacentrum.cz - doom30.metacentrum.cz 2x nVidia Tesla K20 5GB (aka Kepler) 3.5 YES cuda35,cuda20
konos.fav.zcu.cz konos1.fav.zcu.cz - konos10.fav.zcu.cz 2x GPU NVIDIA GeForce GTX 465 2.0 No cuda20
gram.zcu.cz gram1.zcu.cz - gram10.zcu.cz 4x nVidia Tesla M2090 6GB 2.0 No cuda20
zubat.ncbr.muni.cz zubat1.ncbr.muni.cz - zubat8.ncbr.muni.cz 2x nVidia Tesla K20Xm 6GB (aka Kepler) 3.5 YES cuda35,cuda20
glados.cerit-sc.cz glados10.cerit-sc.cz - glados17.cerit-sc.cz Nvidia 1080Ti GPU 3.5 YES cuda35,cuda20


Licence

You have to be registered in NVIDIA Accelerated Computing Developer Program and agree with their licence.

Use

module load cudnn-7.0

To plan your job on clusters with certain Compute Capability, use qsub command like this:

qsub -q gpu -l select=1:ncpus=1:ngpus=X:gpu_cap=cuda35 <job batch file>

Documentation

https://developer.nvidia.com/cudnn

Homepage

https://developer.nvidia.com/cudnn

Program manager

meta@cesnet.cz