Crea il tuo contenitore Docker con la libreria parallela di dati SageMaker distribuiti

Per creare un Dockerfile con il toolkit di SageMaker formazione e la libreria data parallel

1. Inizia con un'immagine Docker di. NVIDIACUDA Usa le versioni per DNN sviluppatori di cu che contengono strumenti CUDA di runtime e di sviluppo (intestazioni e librerie) per creare a partire dal PyTorch codice sorgente.

   FROM nvidia/cuda:11.3.1-cudnn8-devel-ubuntu20.04

   Suggerimento

   Le immagini ufficiali di AWS Deep Learning Container (DLC) sono create a partire dalle immagini di NVIDIACUDAbase. Se desideri utilizzare le DLC immagini predefinite come riferimenti mentre segui il resto delle istruzioni, consulta AWS Deep Learning Containers for PyTorch Dockerfiles.

2. Aggiungi i seguenti argomenti per specificare le versioni PyTorch e gli altri pacchetti. Inoltre, indica i percorsi dei bucket Amazon S3 verso la libreria SageMaker parallela di dati e altri software per utilizzare AWS le risorse, come il plug-in Amazon S3.

   Per utilizzare versioni delle librerie di terze parti diverse da quelle fornite nel seguente esempio di codice, ti consigliamo di consultare i Dockerfile ufficiali di AWS Deep Learning Container PyTorch per trovare versioni testate, compatibili e adatte alla tua applicazione.

   URLsPer trovare l'SMDATAPARALLEL_BINARYargomento, consulta le tabelle di ricerca all'indirizzo. Framework supportati

   ARG PYTORCH_VERSION=1.10.2
   ARG PYTHON_SHORT_VERSION=3.8
   ARG EFA_VERSION=1.14.1
   ARG SMDATAPARALLEL_BINARY=https://smdataparallel.s3.amazonaws.com/binary/pytorch/${PYTORCH_VERSION}/cu113/2022-02-18/smdistributed_dataparallel-1.4.0-cp38-cp38-linux_x86_64.whl
   ARG PT_S3_WHL_GPU=https://aws-s3-plugin.s3.us-west-2.amazonaws.com/binaries/0.0.1/1c3e69e/awsio-0.0.1-cp38-cp38-manylinux1_x86_64.whl
   ARG CONDA_PREFIX="/opt/conda"
   ARG BRANCH_OFI=1.1.3-aws

3. Imposta le seguenti variabili di ambiente per creare correttamente i componenti di SageMaker addestramento ed eseguire la libreria parallela di dati. Queste variabili vengono utilizzate per i componenti nelle fasi successive.

   # Set ENV variables required to build PyTorch
   ENV TORCH_CUDA_ARCH_LIST="7.0+PTX 8.0"
   ENV TORCH_NVCC_FLAGS="-Xfatbin -compress-all"
   ENV NCCL_VERSION=2.10.3
   
   # Add OpenMPI to the path.
   ENV PATH /opt/amazon/openmpi/bin:$PATH
   
   # Add Conda to path
   ENV PATH $CONDA_PREFIX/bin:$PATH
   
   # Set this enviroment variable for SageMaker to launch SMDDP correctly.
   ENV SAGEMAKER_TRAINING_MODULE=sagemaker_pytorch_container.training:main
   
   # Add enviroment variable for processes to be able to call fork()
   ENV RDMAV_FORK_SAFE=1
   
   # Indicate the container type
   ENV DLC_CONTAINER_TYPE=training
   
   # Add EFA and SMDDP to LD library path
   ENV LD_LIBRARY_PATH="/opt/conda/lib/python${PYTHON_SHORT_VERSION}/site-packages/smdistributed/dataparallel/lib:$LD_LIBRARY_PATH"
   ENV LD_LIBRARY_PATH=/opt/amazon/efa/lib/:$LD_LIBRARY_PATH

4. Installa o aggiorna curl, wget e git per scaricare e creare pacchetti nelle fasi successive.

   RUN --mount=type=cache,id=apt-final,target=/var/cache/apt \
       apt-get update && apt-get install -y  --no-install-recommends \
           curl \
           wget \
           git \
       && rm -rf /var/lib/apt/lists/*

5. Installa il software Elastic Fabric Adapter (EFA) per la comunicazione EC2 di rete Amazon.

   RUN DEBIAN_FRONTEND=noninteractive apt-get update
   RUN mkdir /tmp/efa \
       && cd /tmp/efa \
       && curl --silent -O https://efa-installer.amazonaws.com/aws-efa-installer-${EFA_VERSION}.tar.gz \
       && tar -xf aws-efa-installer-${EFA_VERSION}.tar.gz \
       && cd aws-efa-installer \
       && ./efa_installer.sh -y --skip-kmod -g \
       && rm -rf /tmp/efa

6. Installa Conda per eseguire la gestione dei pacchetti.

   RUN curl -fsSL -v -o ~/miniconda.sh -O  https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh  && \
       chmod +x ~/miniconda.sh && \
       ~/miniconda.sh -b -p $CONDA_PREFIX && \
       rm ~/miniconda.sh && \
       $CONDA_PREFIX/bin/conda install -y python=${PYTHON_SHORT_VERSION} conda-build pyyaml numpy ipython && \
       $CONDA_PREFIX/bin/conda clean -ya

7. Scarica, compila e installa PyTorch e le relative dipendenze. Creiamo PyTorch a partire dal codice sorgente perché dobbiamo avere il controllo della NCCL versione per garantire la compatibilità con il AWS OFINCCLplug-in.

   1. Seguendo i passaggi del dockerfile PyTorch ufficiale, installa build dependencies e configura ccache per velocizzare la ricompilazione.

      RUN DEBIAN_FRONTEND=noninteractive \
          apt-get install -y --no-install-recommends \
              build-essential \
              ca-certificates \
              ccache \
              cmake \
              git \
              libjpeg-dev \
              libpng-dev \
          && rm -rf /var/lib/apt/lists/*
        
      # Setup ccache
      RUN /usr/sbin/update-ccache-symlinks
      RUN mkdir /opt/ccache && ccache --set-config=cache_dir=/opt/ccache

   2. Le dipendenze comuni di Install PyTorch e Linux.

      # Common dependencies for PyTorch
      RUN conda install astunparse numpy ninja pyyaml mkl mkl-include setuptools cmake cffi typing_extensions future six requests dataclasses
      
      # Linux specific dependency for PyTorch
      RUN conda install -c pytorch magma-cuda113

   3. Clona il PyTorch GitHub repository.

      RUN --mount=type=cache,target=/opt/ccache \
          cd / \
          && git clone --recursive https://github.com/pytorch/pytorch -b v${PYTORCH_VERSION}

   4. Installa e crea una versione specifica NCCL. Per fare ciò, sostituisci il contenuto nella NCCL cartella predefinita (/pytorch/third_party/nccl) con la NCCL versione specifica del NVIDIA repository. PyTorch La NCCL versione è stata impostata nel passaggio 3 di questa guida.

      RUN cd /pytorch/third_party/nccl \
          && rm -rf nccl \
          && git clone https://github.com/NVIDIA/nccl.git -b v${NCCL_VERSION}-1 \
          && cd nccl \
          && make -j64 src.build CUDA_HOME=/usr/local/cuda NVCC_GENCODE="-gencode=arch=compute_70,code=sm_70 -gencode=arch=compute_80,code=sm_80" \
          && make pkg.txz.build \
          && tar -xvf build/pkg/txz/nccl_*.txz -C $CONDA_PREFIX --strip-components=1

   5. Compila e installa PyTorch. Il completamento di questo processo richiede in genere poco più di 1 ora. Viene creato utilizzando la NCCL versione scaricata in un passaggio precedente.

      RUN cd /pytorch \
          && CMAKE_PREFIX_PATH="$(dirname $(which conda))/../" \
          python setup.py install \
          && rm -rf /pytorch

8. Crea e installa il AWS OFINCCLplugin. Ciò abilita il supporto libfabric per la libreria SageMaker data parallel.

   RUN DEBIAN_FRONTEND=noninteractive apt-get update \
       && apt-get install -y --no-install-recommends \
           autoconf \
           automake \
           libtool
   RUN mkdir /tmp/efa-ofi-nccl \
       && cd /tmp/efa-ofi-nccl \
       && git clone https://github.com/aws/aws-ofi-nccl.git -b v${BRANCH_OFI} \
       && cd aws-ofi-nccl \
       && ./autogen.sh \
       && ./configure --with-libfabric=/opt/amazon/efa \
       --with-mpi=/opt/amazon/openmpi \
       --with-cuda=/usr/local/cuda \
       --with-nccl=$CONDA_PREFIX \
       && make \
       && make install \
       && rm -rf /tmp/efa-ofi-nccl

9. Compila e installa TorchVision.

   RUN pip install --no-cache-dir -U \
       packaging \
       mpi4py==3.0.3
   RUN cd /tmp \
       && git clone https://github.com/pytorch/vision.git -b v0.9.1 \
       && cd vision \
       && BUILD_VERSION="0.9.1+cu111" python setup.py install \
       && cd /tmp \
       && rm -rf vision

10. Installa e configura OpenSSH. Open SSH è necessario MPI per comunicare tra contenitori. Consenti SSH a Open di parlare con i container senza chiedere conferma.

    RUN apt-get update \
        && apt-get install -y  --allow-downgrades --allow-change-held-packages --no-install-recommends \
        && apt-get install -y --no-install-recommends openssh-client openssh-server \
        && mkdir -p /var/run/sshd \
        && cat /etc/ssh/ssh_config | grep -v StrictHostKeyChecking > /etc/ssh/ssh_config.new \
        && echo "    StrictHostKeyChecking no" >> /etc/ssh/ssh_config.new \
        && mv /etc/ssh/ssh_config.new /etc/ssh/ssh_config \
        && rm -rf /var/lib/apt/lists/*
    
    # Configure OpenSSH so that nodes can communicate with each other
    RUN mkdir -p /var/run/sshd && \
     sed 's@session\s*required\s*pam_loginuid.so@session optional pam_loginuid.so@g' -i /etc/pam.d/sshd
    RUN rm -rf /root/.ssh/ && \
     mkdir -p /root/.ssh/ && \
     ssh-keygen -q -t rsa -N '' -f /root/.ssh/id_rsa && \
     cp /root/.ssh/id_rsa.pub /root/.ssh/authorized_keys \
     && printf "Host *\n StrictHostKeyChecking no\n" >> /root/.ssh/config

11. Installa il plug-in PT S3 per accedere in modo efficiente ai set di dati in Amazon S3.

    RUN pip install --no-cache-dir -U ${PT_S3_WHL_GPU}
    RUN mkdir -p /etc/pki/tls/certs && cp /etc/ssl/certs/ca-certificates.crt /etc/pki/tls/certs/ca-bundle.crt

12. Installa la libreria libboost. Questo pacchetto è necessario per collegare in rete la funzionalità IO asincrona della libreria data SageMaker parallel.

    WORKDIR /
    RUN wget https://sourceforge.net/projects/boost/files/boost/1.73.0/boost_1_73_0.tar.gz/download -O boost_1_73_0.tar.gz \
        && tar -xzf boost_1_73_0.tar.gz \
        && cd boost_1_73_0 \
        && ./bootstrap.sh \
        && ./b2 threading=multi --prefix=${CONDA_PREFIX} -j 64 cxxflags=-fPIC cflags=-fPIC install || true \
        && cd .. \
        && rm -rf boost_1_73_0.tar.gz \
        && rm -rf boost_1_73_0 \
        && cd ${CONDA_PREFIX}/include/boost

13. Installa i seguenti SageMaker strumenti per la formazione. PyTorch

    WORKDIR /root
    RUN pip install --no-cache-dir -U \
        smclarify \
        "sagemaker>=2,<3" \
        sagemaker-experiments==0.* \
        sagemaker-pytorch-training

14. Infine, installa il binario SageMaker data parallel e le dipendenze rimanenti.

    RUN --mount=type=cache,id=apt-final,target=/var/cache/apt \
      apt-get update && apt-get install -y  --no-install-recommends \
      jq \
      libhwloc-dev \
      libnuma1 \
      libnuma-dev \
      libssl1.1 \
      libtool \
      hwloc \
      && rm -rf /var/lib/apt/lists/*
    
    RUN SMDATAPARALLEL_PT=1 pip install --no-cache-dir ${SMDATAPARALLEL_BINARY}

15. Dopo aver finito di creare il Dockerfile, consulta Adapting Your Own Training Container per scoprire come creare il contenitore Docker, ospitarlo in Amazon ed eseguire un ECR processo di formazione utilizzando Python. SageMaker SDK

# This file creates a docker image with minimum dependencies to run SageMaker data parallel training
FROM nvidia/cuda:11.3.1-cudnn8-devel-ubuntu20.04

# Set appropiate versions and location for components
ARG PYTORCH_VERSION=1.10.2
ARG PYTHON_SHORT_VERSION=3.8
ARG EFA_VERSION=1.14.1
ARG SMDATAPARALLEL_BINARY=https://smdataparallel.s3.amazonaws.com/binary/pytorch/${PYTORCH_VERSION}/cu113/2022-02-18/smdistributed_dataparallel-1.4.0-cp38-cp38-linux_x86_64.whl
ARG PT_S3_WHL_GPU=https://aws-s3-plugin.s3.us-west-2.amazonaws.com/binaries/0.0.1/1c3e69e/awsio-0.0.1-cp38-cp38-manylinux1_x86_64.whl
ARG CONDA_PREFIX="/opt/conda"
ARG BRANCH_OFI=1.1.3-aws

# Set ENV variables required to build PyTorch
ENV TORCH_CUDA_ARCH_LIST="3.7 5.0 7.0+PTX 8.0"
ENV TORCH_NVCC_FLAGS="-Xfatbin -compress-all"
ENV NCCL_VERSION=2.10.3

# Add OpenMPI to the path.
ENV PATH /opt/amazon/openmpi/bin:$PATH

# Add Conda to path
ENV PATH $CONDA_PREFIX/bin:$PATH

# Set this enviroment variable for SageMaker to launch SMDDP correctly.
ENV SAGEMAKER_TRAINING_MODULE=sagemaker_pytorch_container.training:main

# Add enviroment variable for processes to be able to call fork()
ENV RDMAV_FORK_SAFE=1

# Indicate the container type
ENV DLC_CONTAINER_TYPE=training

# Add EFA and SMDDP to LD library path
ENV LD_LIBRARY_PATH="/opt/conda/lib/python${PYTHON_SHORT_VERSION}/site-packages/smdistributed/dataparallel/lib:$LD_LIBRARY_PATH"
ENV LD_LIBRARY_PATH=/opt/amazon/efa/lib/:$LD_LIBRARY_PATH

# Install basic dependencies to download and build other dependencies
RUN --mount=type=cache,id=apt-final,target=/var/cache/apt \
  apt-get update && apt-get install -y  --no-install-recommends \
  curl \
  wget \
  git \
  && rm -rf /var/lib/apt/lists/*

# Install EFA.
# This is required for SMDDP backend communication
RUN DEBIAN_FRONTEND=noninteractive apt-get update
RUN mkdir /tmp/efa \
    && cd /tmp/efa \
    && curl --silent -O https://efa-installer.amazonaws.com/aws-efa-installer-${EFA_VERSION}.tar.gz \
    && tar -xf aws-efa-installer-${EFA_VERSION}.tar.gz \
    && cd aws-efa-installer \
    && ./efa_installer.sh -y --skip-kmod -g \
    && rm -rf /tmp/efa

# Install Conda
RUN curl -fsSL -v -o ~/miniconda.sh -O  https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh  && \
    chmod +x ~/miniconda.sh && \
    ~/miniconda.sh -b -p $CONDA_PREFIX && \
    rm ~/miniconda.sh && \
    $CONDA_PREFIX/bin/conda install -y python=${PYTHON_SHORT_VERSION} conda-build pyyaml numpy ipython && \
    $CONDA_PREFIX/bin/conda clean -ya

# Install PyTorch.
# Start with dependencies listed in official PyTorch dockerfile
# https://github.com/pytorch/pytorch/blob/master/Dockerfile
RUN DEBIAN_FRONTEND=noninteractive \
    apt-get install -y --no-install-recommends \
        build-essential \
        ca-certificates \
        ccache \
        cmake \
        git \
        libjpeg-dev \
        libpng-dev && \
    rm -rf /var/lib/apt/lists/*

# Setup ccache
RUN /usr/sbin/update-ccache-symlinks
RUN mkdir /opt/ccache && ccache --set-config=cache_dir=/opt/ccache

# Common dependencies for PyTorch
RUN conda install astunparse numpy ninja pyyaml mkl mkl-include setuptools cmake cffi typing_extensions future six requests dataclasses

# Linux specific dependency for PyTorch
RUN conda install -c pytorch magma-cuda113

# Clone PyTorch
RUN --mount=type=cache,target=/opt/ccache \
    cd / \
    && git clone --recursive https://github.com/pytorch/pytorch -b v${PYTORCH_VERSION}
# Note that we need to use the same NCCL version for PyTorch and OFI plugin.
# To enforce that, install NCCL from source before building PT and OFI plugin.

# Install NCCL.
# Required for building OFI plugin (OFI requires NCCL's header files and library)
RUN cd /pytorch/third_party/nccl \
    && rm -rf nccl \
    && git clone https://github.com/NVIDIA/nccl.git -b v${NCCL_VERSION}-1 \
    && cd nccl \
    && make -j64 src.build CUDA_HOME=/usr/local/cuda NVCC_GENCODE="-gencode=arch=compute_70,code=sm_70 -gencode=arch=compute_80,code=sm_80" \
    && make pkg.txz.build \
    && tar -xvf build/pkg/txz/nccl_*.txz -C $CONDA_PREFIX --strip-components=1

# Build and install PyTorch.
RUN cd /pytorch \
    && CMAKE_PREFIX_PATH="$(dirname $(which conda))/../" \
    python setup.py install \
    && rm -rf /pytorch

RUN ccache -C

# Build and install OFI plugin. \
# It is required to use libfabric.
RUN DEBIAN_FRONTEND=noninteractive apt-get update \
    && apt-get install -y --no-install-recommends \
        autoconf \
        automake \
        libtool
RUN mkdir /tmp/efa-ofi-nccl \
    && cd /tmp/efa-ofi-nccl \
    && git clone https://github.com/aws/aws-ofi-nccl.git -b v${BRANCH_OFI} \
    && cd aws-ofi-nccl \
    && ./autogen.sh \
    && ./configure --with-libfabric=/opt/amazon/efa \
        --with-mpi=/opt/amazon/openmpi \
        --with-cuda=/usr/local/cuda \
        --with-nccl=$CONDA_PREFIX \
    && make \
    && make install \
    && rm -rf /tmp/efa-ofi-nccl

# Build and install Torchvision
RUN pip install --no-cache-dir -U \
    packaging \
    mpi4py==3.0.3
RUN cd /tmp \
    && git clone https://github.com/pytorch/vision.git -b v0.9.1 \
    && cd vision \
    && BUILD_VERSION="0.9.1+cu111" python setup.py install \
    && cd /tmp \
    && rm -rf vision

# Install OpenSSH.
# Required for MPI to communicate between containers, allow OpenSSH to talk to containers without asking for confirmation
RUN apt-get update \
    && apt-get install -y  --allow-downgrades --allow-change-held-packages --no-install-recommends \
    && apt-get install -y --no-install-recommends openssh-client openssh-server \
    && mkdir -p /var/run/sshd \
    && cat /etc/ssh/ssh_config | grep -v StrictHostKeyChecking > /etc/ssh/ssh_config.new \
    && echo "    StrictHostKeyChecking no" >> /etc/ssh/ssh_config.new \
    && mv /etc/ssh/ssh_config.new /etc/ssh/ssh_config \
    && rm -rf /var/lib/apt/lists/*
# Configure OpenSSH so that nodes can communicate with each other
RUN mkdir -p /var/run/sshd && \
    sed 's@session\s*required\s*pam_loginuid.so@session optional pam_loginuid.so@g' -i /etc/pam.d/sshd
RUN rm -rf /root/.ssh/ && \
    mkdir -p /root/.ssh/ && \
    ssh-keygen -q -t rsa -N '' -f /root/.ssh/id_rsa && \
    cp /root/.ssh/id_rsa.pub /root/.ssh/authorized_keys \
    && printf "Host *\n StrictHostKeyChecking no\n" >> /root/.ssh/config

# Install PT S3 plugin.
# Required to efficiently access datasets in Amazon S3
RUN pip install --no-cache-dir -U ${PT_S3_WHL_GPU}
RUN mkdir -p /etc/pki/tls/certs && cp /etc/ssl/certs/ca-certificates.crt /etc/pki/tls/certs/ca-bundle.crt

# Install libboost from source.
# This package is needed for smdataparallel functionality (for networking asynchronous IO).
WORKDIR /
RUN wget https://sourceforge.net/projects/boost/files/boost/1.73.0/boost_1_73_0.tar.gz/download -O boost_1_73_0.tar.gz \
    && tar -xzf boost_1_73_0.tar.gz \
    && cd boost_1_73_0 \
    && ./bootstrap.sh \
    && ./b2 threading=multi --prefix=${CONDA_PREFIX} -j 64 cxxflags=-fPIC cflags=-fPIC install || true \
    && cd .. \
    && rm -rf boost_1_73_0.tar.gz \
    && rm -rf boost_1_73_0 \
    && cd ${CONDA_PREFIX}/include/boost

# Install SageMaker PyTorch training.
WORKDIR /root
RUN pip install --no-cache-dir -U \
    smclarify \
    "sagemaker>=2,<3" \
    sagemaker-experiments==0.* \
    sagemaker-pytorch-training

# Install SageMaker data parallel binary (SMDDP)
# Start with dependencies
RUN --mount=type=cache,id=apt-final,target=/var/cache/apt \
    apt-get update && apt-get install -y  --no-install-recommends \
        jq \
        libhwloc-dev \
        libnuma1 \
        libnuma-dev \
        libssl1.1 \
        libtool \
        hwloc \
    && rm -rf /var/lib/apt/lists/*

# Install SMDDP
RUN SMDATAPARALLEL_PT=1 pip install --no-cache-dir ${SMDATAPARALLEL_BINARY}