Cheat Sheet for Singularity (and Docker) recipe

singularity_cheat_sheet.md

cheat sheet for different container snippets

cuda docker sources

Sometimes the PATH and LD_LIBRARY_PATH are not set correctly for the CUDA installation

	export PATH=$PATH:/usr/local/cuda/bin/
	export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/cuda/lib6

• x86: https://hub.docker.com/r/nvidia/cuda
  • Tags:
    • nvidia/cuda:8.0-devel-ubuntu16.04
    • nvidia/cuda:9.0-devel-ubuntu18.04
    • nvidia/cuda:9.1-devel-ubuntu18.04
    • nvidia/cuda:10.0-devel-ubuntu18.04
    • nvidia/cuda:10.1-devel-ubuntu18.04
• power: https://hub.docker.com/r/nvidia/cuda-ppc64le/tags
  • Tags:
    • nvidia/cuda-ppc64le:10.1-cudnn7-devel-ubuntu18.04

Ubuntu tags

• Ubuntu 16.04
  • ubuntu:xenial
• Ubuntu 18.04
  • ubuntu:bionic

install clang via apt

	wget http://llvm.org/apt/llvm-snapshot.gpg.key
	apt-key add llvm-snapshot.gpg.key
	rm llvm-snapshot.gpg.key

	echo "" >> /etc/apt/sources.list
	echo "deb http://apt.llvm.org/bionic/ llvm-toolchain-bionic main" >> /etc/apt/sources.list
	echo "deb-src http://apt.llvm.org/bionic/ llvm-toolchain-bionic main" >> /etc/apt/sources.list

	apt update
	apt install -y clang-5.0 llvm-5.0 clang-6.0 llvm-6.0 clang-7 llvm-7 clang-8 llvm-8 clang-9 llvm-9

install boost via apt

	add-apt-repository ppa:mhier/libboost-latest -y
	apt-get update
	apt-get install -y boost1.67

using spack

	%environment
		export PATH=/opt/spack/bin:$PATH

	%post
        cd /opt
        git clone https://github.com/spack/spack.git
        export PATH=/opt/spack/bin:$PATH

        spack compiler find
        spack bootstrap
        spack install

using pip and numba

	%environment
        export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/cuda-9.0/lib64
        export NUMBAPRO_NVVM=/usr/local/cuda-9.0/nvvm/lib64/libnvvm.so
        export NUMBAPRO_LIBDEVICE=/usr/local/cuda-9.0/nvvm/libdevice
        export NUMBAPRO_CUDA_DRIVER=/.singularity.d/libs/libcuda.so

	%post
        echo 'install additional software'
        apt update
        apt install -y nano python3 python3-pip
        python3 -m pip install virtualenv numba