NVIDIA
NVIDIA
CUDA Deep Learning
Container
NVIDIA
NVIDIA
CUDA Deep Learning

CUDA is a parallel computing platform and programming model that enhances computing performance using NVIDIA GPUs. CUDA Deep Learning integrates networking and GPU-accelerated libraries like cuDNN, cuTensor, NCCL, HPC-x, and the CUDA Toolkit.

LayerLabelCreated
b20dc2b29db077d451c6126b263d6698bd5d5b2338f5bc2975cf8e58c3f29475CONFIG
Entrypoint /opt/nvidia/nvidia_entrypoint.sh
02/19/2025 8:36 PM UTC
f8a97d5b44f76b5e261d903c622305ddd2a442ffdb62a5ef4bbb3020f8f1336bCOPY
/opt/amazon/aws-ofi-nccl /opt/amazon/aws-ofi-nccl
02/19/2025 8:36 PM UTC
ccc582fa9a8471aed8573d2e3a6c1f47deb29142d93b4a672ced7e8cc3b3b126ENV
LIBRARY_PATH=/usr/local/cuda/lib64/stubs:
02/19/2025 8:36 PM UTC
e491cb646daa053c944b38f050f0632416199d1279a86a4a113895104eeb7c42RUN
GDRCOPY_VERSION=2.4.1 HPCX_VERSION=2.22.1 RDMACORE_VERSION=50.0 MOFED_VERSION=5.4-rdmacore50.0 OPENUCX_VERSION=1.18.0 OPENMPI_VERSION=4.1.7 EFA_VERSION=1.34.0 AWS_OFI_NCCL_VERSION=1.12.1 TARGETARCH=amd64 export DEVEL=1 BASE=0 &&
  /nvidia/build-scripts/installNCU.sh &&
  /nvidia/build-scripts/installCUDA.sh &&
  /nvidia/build-scripts/installLIBS.sh &&
  if [ ! -f /etc/ld.so.conf.d/nvidia-tegra.conf ]; then /nvidia/build-scripts/installNCCL.sh; fi &&
  /nvidia/build-scripts/installCUDNN.sh &&
  /nvidia/build-scripts/installCUTENSOR.sh &&
  /nvidia/build-scripts/installTRT.sh &&
  /nvidia/build-scripts/installNSYS.sh &&
  /nvidia/build-scripts/installCUSPARSELT.sh &&
  if [ -f "/tmp/cuda-${_CUDA_VERSION_MAJMIN}.patch" ]; then patch -p0 < /tmp/cuda-${_CUDA_VERSION_MAJMIN}.patch; fi &&
  rm -f /tmp/cuda-*.patch
02/19/2025 8:36 PM UTC
9bbe90cc5062fad86d10ae38291a73655b03c2db47d75be81aa2bcf3bff5651fCOPY
cuda-*.patch /tmp
02/19/2025 8:33 PM UTC
f9fa93611d8b08f28cd9ba6d7caa455b307235d774d126d4c1d02a92579fac9cENV
OMPI_MCA_coll_hcoll_enable=0
02/19/2025 8:33 PM UTC
7cd101da0ea865c7b5817d40d1419ce947668de4df1376f6a890d7867cb1923bENV
OPAL_PREFIX=/opt/hpcx/ompi PATH=/usr/local/mpi/bin:/usr/local/nvidia/bin:/usr/local/cuda/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/local/ucx/bin:/opt/amazon/efa/bin
02/19/2025 8:33 PM UTC
d760631ccdba1e07aa0f90cb3da52785ad96b5fa41dcab710d1a9cec5e91e72cRUN
GDRCOPY_VERSION=2.4.1 HPCX_VERSION=2.22.1 RDMACORE_VERSION=50.0 MOFED_VERSION=5.4-rdmacore50.0 OPENUCX_VERSION=1.18.0 OPENMPI_VERSION=4.1.7 EFA_VERSION=1.34.0 AWS_OFI_NCCL_VERSION=1.12.1 TARGETARCH=amd64 cd /nvidia &&
  ( export DEBIAN_FRONTEND=noninteractive &&
  apt-get update &&
  apt-get install -y --no-install-recommends libibverbs1 libibverbs-dev librdmacm1 librdmacm-dev libibumad3 libibumad-dev ibverbs-utils ibverbs-providers &&
  rm -rf /var/lib/apt/lists/* &&
  rm $(dpkg-query -L libibverbs-dev librdmacm-dev libibumad-dev | grep "\(\.so\|\.a\)$") ) &&
  ( cd opt/gdrcopy/ &&
  dpkg -i libgdrapi_*.deb ) &&
  ( cp -r opt/hpcx /opt/ &&
  cp etc/ld.so.conf.d/hpcx.conf /etc/ld.so.conf.d/ &&
  ln -sf /opt/hpcx/ompi /usr/local/mpi &&
  ln -sf /opt/hpcx/ucx /usr/local/ucx &&
  sed -i 's/^\(hwloc_base_binding_policy\) = core$/\1 = none/' /opt/hpcx/ompi/etc/openmpi-mca-params.conf &&
  sed -i 's/^\(btl = self\)$/#\1/' /opt/hpcx/ompi/etc/openmpi-mca-params.conf ) &&
  ( if [ ! -f /etc/ld.so.conf.d/nvidia-tegra.conf ]; then cd opt/amazon/efa/ &&
  dpkg -i libfabric*.deb &&
  rm /opt/amazon/efa/lib/libfabric.a &&
  echo "/opt/amazon/efa/lib" > /etc/ld.so.conf.d/efa.conf; fi ) &&
  ldconfig
02/19/2025 8:33 PM UTC
89a5fc6f3f62468d69cadd2866039705bd4b05da1d3da2c68e524e103e10676eARG
TARGETARCH=amd64
02/19/2025 8:33 PM UTC
8aa0b463eee1f24ac5365f7bdfb9ebfb19eb70acbc4d60f3ada987d6d7558ea1ENV
GDRCOPY_VERSION=2.4.1 HPCX_VERSION=2.22.1 MOFED_VERSION=5.4-rdmacore50.0 OPENUCX_VERSION=1.18.0 OPENMPI_VERSION=4.1.7 RDMACORE_VERSION=50.0 EFA_VERSION=1.34.0 AWS_OFI_NCCL_VERSION=1.12.1
02/19/2025 8:33 PM UTC
...