NCCL 2.4.8-gcccuda-2019b

Unsupported: Use of this version of NCCL is not supported. More information on our Applications Support and Retention Policy.

There is a newer version of NCCL

The NVIDIA Collective Communications Library (NCCL) implements multi-GPU and multi-node collective communication primitives that are performance optimized for NVIDIA GPUs.

Accessing NCCL 2.4.8-gcccuda-2019b

To load the module for NCCL 2.4.8-gcccuda-2019b please use this command on the BEAR systems (BlueBEAR, BEAR Cloud VMs, and CaStLeS VMs):

module load bear-apps-unsupported/handbuilt/2019
module load NCCL/2.4.8-gcccuda-2019b

BEAR Apps Version

2019h

More Information

For more information visit the NCCL website.

Dependencies

This version of NCCL has a direct dependency on: gcccuda/2019b

Required By

This version of NCCL is a direct dependent of: libgpuarray/0.7.6-fosscuda-2019b-Python-3.7.4 PyTorch/1.4.0-fosscuda-2019b-Python-3.7.4 PyTorch/1.3.1-fosscuda-2019b-Python-3.7.4 PyTorch/1.6.0-fosscuda-2019b-Python-3.7.4 TensorFlow/1.15.0-fosscuda-2019b-Python-3.7.4 TensorFlow/2.0.0-fosscuda-2019b-Python-3.7.4 TensorFlow/2.2.0-fosscuda-2019b-Python-3.7.4

Other Versions

These versions of NCCL are available on the BEAR systems (BlueBEAR, BEAR Cloud VMs, and CaStLeS VMs). These will be retained in accordance with our Applications Support and Retention Policy.

Last modified on 19th November 2019