Add Nvidia Metropolis Microservices Essentials package to Photon OS

[dcasota]

Is your feature request related to a problem? Please describe.

Please consider a build-essential like Nvidia Metropolis Microservices "Essential" package as Nvidia GPUs are very popular in Broadcom's co-packaged VMware products.

Describe the solution you'd like

tdnf install NvidiaMM-essential installs and configures Nvidia drivers, docker, Nvidia container toolkit, Ngc cli.

Easy consumerization of packages/containers Tensorflow, Pytorch, tensorRT, Triton Inference Server, Cuda, NVCaffe, Tao Toolkit, Riva, Nemo, Clara, etc.

Describe alternatives you've considered

Old (proprietary) labs e.g. Video Processing on Photon OS-sig.pdf will become easier maintainable because data scientists + AI students do not need insights of Nvidia Metropolis Microservices Essentials on Photon OS.

Additional context

• nvidia container toolkit

  # Since 1.14, spec file https://github.com/NVIDIA/libnvidia-container/blob/main/pkg/rpm/SPECS/libnvidia-container.spec
  sudo tdnf install -y curl gpg
  sudo curl -fsSL https://nvidia.github.io/libnvidia-container/gpgkey | sudo gpg --dearmor -o /etc/pki/rpm-gpg/nvidia-container-toolkit-keyring.gpg
  sudo curl -s -L https://nvidia.github.io/libnvidia-container/stable/rpm/nvidia-container-toolkit.repo |sudo tee /etc/yum.repos.d/nvidia-container-toolkit.repo
  sudo tdnf makecache
  sudo tdnf install -y nvidia-container-toolkit nvidia-docker2
  sudo systemctl restart docker
  

• ngc-cli

  sudo tdnf install -y unzip
  cd /usr/local/bin
  sudo curl -J -L -O https://ngc.nvidia.com/downloads/ngccli_cat_linux.zip
  sudo unzip ngccli_cat_linux.zip
  sudo chmod u+x ngc-cli/ngc
  sudo rm ngccli_cat_linux.zip ngc-cli.md5
  echo "no-apikey\nascii\n" | sudo ngc-cli/ngc config set
  

[dcasota]

No answer -> not planned / no priority-> close ticket.