Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
241 changes: 241 additions & 0 deletions jax/training/0.10.0/Dockerfile.neuronx
Original file line number Diff line number Diff line change
@@ -0,0 +1,241 @@
ARG BUILD_STAGE=prod

FROM public.ecr.aws/docker/library/ubuntu:24.04 AS base

LABEL dlc_major_version="1"
LABEL maintainer="Amazon AI"

# This arg required to stop docker build waiting for region configuration while installing tz data from ubuntu 24
ARG DEBIAN_FRONTEND=noninteractive
ARG PYTHON=python3.13
ARG PYTHON_VERSION=3.13.13
ARG PIP=pip3
ARG OMPI_VERSION=4.1.5
ARG PYPI_SIMPLE_URL="https://pypi.org/simple/"

# Python won’t try to write .pyc or .pyo files on the import of source modules
# Force stdin, stdout and stderr to be totally unbuffered. Good for logging
ENV PYTHONDONTWRITEBYTECODE=1
ENV PYTHONUNBUFFERED=1
ENV PYTHONIOENCODING=UTF-8
ENV LANG=C.UTF-8
ENV LC_ALL=C.UTF-8
ENV LD_LIBRARY_PATH="${LD_LIBRARY_PATH}:/opt/aws/neuron/lib"

Check warning on line 23 in jax/training/0.10.0/Dockerfile.neuronx

View workflow job for this annotation

GitHub Actions / build (0.10.0, jax/training/0.10.0)

Variables should be defined before their use

UndefinedVar: Usage of undefined variable '$LD_LIBRARY_PATH' More info: https://docs.docker.com/go/dockerfile/rule/undefined-var/
ENV LD_LIBRARY_PATH="${LD_LIBRARY_PATH}:/opt/amazon/efa/lib"
ENV LD_LIBRARY_PATH="${LD_LIBRARY_PATH}:/opt/amazon/efa/lib64"
ENV LD_LIBRARY_PATH="${LD_LIBRARY_PATH}:/opt/amazon/openmpi/lib64"
ENV LD_LIBRARY_PATH="${LD_LIBRARY_PATH}:/usr/local/lib"
ENV PATH="/opt/aws/neuron/bin:${PATH}"
ENV PATH="${PATH}:/opt/amazon/efa/bin"

RUN apt-get update \

Check failure on line 31 in jax/training/0.10.0/Dockerfile.neuronx

View workflow job for this annotation

GitHub Actions / dockerfile-linter

DL3008 warning: Pin versions in apt get install. Instead of `apt-get install <package>` use `apt-get install <package>=<version>`
&& apt-get upgrade -y \
&& apt-get install -y --no-install-recommends \
build-essential \
ca-certificates \
cmake \
curl \
git \
gnupg2 \
gpg-agent \
jq \
libglib2.0-0 \
libgl1-mesa-dri \
libsm6 \
libxext6 \
libxrender-dev \
libssl-dev \
libsqlite3-dev \
libgdbm-dev \
libc6-dev \
libbz2-dev \
libncurses-dev \
libffi-dev \
libcap-dev \
libhwloc-dev \
openjdk-8-jdk-headless \
openjdk-8-jdk \
openjdk-8-jre \
openjdk-11-jdk \
openssl \
software-properties-common \
tk-dev \
unzip \
wget \
zlib1g-dev \
&& rm -rf /var/lib/apt/lists/* \
&& rm -rf /tmp/tmp* \
&& apt-get clean

# Install Open MPI and configure SSH for MPI operator in k8s
RUN mkdir -p /tmp/openmpi \

Check failure on line 71 in jax/training/0.10.0/Dockerfile.neuronx

View workflow job for this annotation

GitHub Actions / dockerfile-linter

SC2046 warning: Quote this to prevent word splitting.

Check failure on line 71 in jax/training/0.10.0/Dockerfile.neuronx

View workflow job for this annotation

GitHub Actions / dockerfile-linter

DL3003 warning: Use WORKDIR to switch to a directory
&& cd /tmp/openmpi \
&& wget --quiet https://download.open-mpi.org/release/open-mpi/v4.1/openmpi-${OMPI_VERSION}.tar.gz \
&& tar zxf openmpi-${OMPI_VERSION}.tar.gz \
&& cd openmpi-${OMPI_VERSION} \
&& ./configure --enable-orterun-prefix-by-default \
&& make -j $(nproc) all \
&& make install \
&& ldconfig \
&& rm -rf /tmp/openmpi

# Install packages and configure SSH for MPI operator in k8s
RUN apt-get update \

Check failure on line 83 in jax/training/0.10.0/Dockerfile.neuronx

View workflow job for this annotation

GitHub Actions / dockerfile-linter

DL3015 info: Avoid additional packages by specifying `--no-install-recommends`

Check failure on line 83 in jax/training/0.10.0/Dockerfile.neuronx

View workflow job for this annotation

GitHub Actions / dockerfile-linter

DL3008 warning: Pin versions in apt get install. Instead of `apt-get install <package>` use `apt-get install <package>=<version>`
&& apt-get install -y openmpi-bin openssh-server \
&& mkdir -p /var/run/sshd \
&& echo " UserKnownHostsFile /dev/null" >> /etc/ssh/ssh_config \
&& echo " StrictHostKeyChecking no" >> /etc/ssh/ssh_config \
&& sed -i 's/#\(StrictModes \).*/\1no/g' /etc/ssh/sshd_config \
&& rm -rf /var/lib/apt/lists/* \
&& rm -rf /tmp/tmp* \
&& apt-get clean

# Install Python
RUN wget -q https://www.python.org/ftp/python/$PYTHON_VERSION/Python-$PYTHON_VERSION.tgz \

Check failure on line 94 in jax/training/0.10.0/Dockerfile.neuronx

View workflow job for this annotation

GitHub Actions / dockerfile-linter

SC2046 warning: Quote this to prevent word splitting.

Check failure on line 94 in jax/training/0.10.0/Dockerfile.neuronx

View workflow job for this annotation

GitHub Actions / dockerfile-linter

DL3003 warning: Use WORKDIR to switch to a directory
&& tar -xzf Python-$PYTHON_VERSION.tgz \
&& cd Python-$PYTHON_VERSION \
&& ./configure --enable-shared --prefix=/usr/local \
&& make -j $(nproc) && make install \
&& cd .. && rm -rf ../Python-$PYTHON_VERSION* \
&& ln -s /usr/local/bin/pip3 /usr/bin/pip \
&& ln -s /usr/local/bin/$PYTHON /usr/local/bin/python \
&& ${PIP} --no-cache-dir install --upgrade \
pip \
requests \
setuptools \
uv \
&& rm -rf ~/.cache/pip/*

# Install AWS CLI
RUN curl "https://awscli.amazonaws.com/awscli-exe-linux-x86_64.zip" -o "awscliv2.zip" \

Check failure on line 110 in jax/training/0.10.0/Dockerfile.neuronx

View workflow job for this annotation

GitHub Actions / dockerfile-linter

DL4001 warning: Either use Wget or Curl but not both
&& unzip awscliv2.zip \
&& ./aws/install \
&& rm -rf awscliv2.zip aws

# U24 will not allow installation of pip packages outside of venv without this flag
# This is because U24 ships with Python 3.12 by default and installation into the Python
# interpreter’s directory are disabled outside of a virtual environment.
# https://peps.python.org/pep-0668/
RUN ${PIP} config set global.break-system-packages true

# Install EFA
RUN apt-get update \

Check failure on line 122 in jax/training/0.10.0/Dockerfile.neuronx

View workflow job for this annotation

GitHub Actions / dockerfile-linter

DL3047 info: Avoid use of wget without progress bar. Use `wget --progress=dot:giga <url>`. Or consider using `-q` or `-nv` (shorthands for `--quiet` or `--no-verbose`).

Check failure on line 122 in jax/training/0.10.0/Dockerfile.neuronx

View workflow job for this annotation

GitHub Actions / dockerfile-linter

DL4006 warning: Set the SHELL option -o pipefail before RUN with a pipe in it. If you are using /bin/sh in an alpine image or if your shell is symlinked to busybox then consider explicitly setting your SHELL to /bin/ash, or disable this check
&& cd $HOME \
&& curl -O https://efa-installer.amazonaws.com/aws-efa-installer-latest.tar.gz \
&& wget https://efa-installer.amazonaws.com/aws-efa-installer.key && gpg --import aws-efa-installer.key \
&& cat aws-efa-installer.key | gpg --fingerprint \
&& wget https://efa-installer.amazonaws.com/aws-efa-installer-latest.tar.gz.sig && gpg --verify ./aws-efa-installer-latest.tar.gz.sig \
&& tar -xf aws-efa-installer-latest.tar.gz \
&& cd aws-efa-installer \
&& ./efa_installer.sh -y -g --skip-kmod --skip-limit-conf --no-verify \
&& cd $HOME \
&& rm -rf /var/lib/apt/lists/* \
&& rm -rf /tmp/tmp* \
&& apt-get clean

WORKDIR /

# The ENV variables declared below are changed in the previous section
# Grouping these ENV variables in the first section causes
# ompi_info to fail. This is only observed in CPU containers
ENV PATH="$PATH:/home/.openmpi/bin"
ENV LD_LIBRARY_PATH="$LD_LIBRARY_PATH:/home/.openmpi/lib/"
RUN ompi_info --parsable --all | grep mpi_built_with_cuda_support:value

RUN mkdir -p /etc/pki/tls/certs && cp /etc/ssl/certs/ca-certificates.crt /etc/pki/tls/certs/ca-bundle.crt

# Copy workaround script for incorrect hostname
COPY changehostname.c /
COPY --chmod=755 start_with_right_hostname.sh deep_learning_container.py /usr/local/bin/

RUN HOME_DIR=/root \
&& curl -o ${HOME_DIR}/oss_compliance.zip https://aws-dlinfra-utilities.s3.amazonaws.com/oss_compliance.zip \
&& unzip ${HOME_DIR}/oss_compliance.zip -d ${HOME_DIR}/ \
&& cp ${HOME_DIR}/oss_compliance/test/testOSSCompliance /usr/local/bin/testOSSCompliance \
&& chmod +x /usr/local/bin/testOSSCompliance \
&& chmod +x ${HOME_DIR}/oss_compliance/generate_oss_compliance.sh \
&& ${HOME_DIR}/oss_compliance/generate_oss_compliance.sh ${HOME_DIR} ${PYTHON} \
&& rm -rf ${HOME_DIR}/oss_compliance* \
&& rm -rf /tmp/tmp*

# Setting up APT and PIP repo for neuron artifacts
ARG NEURON_APT_REPO=apt.repos.neuron.amazonaws.com
ARG NEURON_APT_REPO_KEY

Check warning on line 163 in jax/training/0.10.0/Dockerfile.neuronx

View workflow job for this annotation

GitHub Actions / build (0.10.0, jax/training/0.10.0)

Sensitive data should not be used in the ARG or ENV commands

SecretsUsedInArgOrEnv: Do not use ARG or ENV instructions for sensitive data (ARG "NEURON_APT_REPO_KEY") More info: https://docs.docker.com/go/dockerfile/rule/secrets-used-in-arg-or-env/
ARG NEURON_PIP_REPO=pip.repos.neuron.amazonaws.com
ARG NEURON_PIP_REPO_KEY

Check warning on line 165 in jax/training/0.10.0/Dockerfile.neuronx

View workflow job for this annotation

GitHub Actions / build (0.10.0, jax/training/0.10.0)

Sensitive data should not be used in the ARG or ENV commands

SecretsUsedInArgOrEnv: Do not use ARG or ENV instructions for sensitive data (ARG "NEURON_PIP_REPO_KEY") More info: https://docs.docker.com/go/dockerfile/rule/secrets-used-in-arg-or-env/
RUN mkdir -p /etc/apt/keyrings \
&& APT_REPO_PREFIX=$([ -n "${NEURON_APT_REPO_KEY}" ] && echo "${NEURON_APT_REPO_KEY}@" || echo "") \
&& echo "deb [signed-by=/etc/apt/keyrings/neuron.gpg] https://${APT_REPO_PREFIX}${NEURON_APT_REPO} jammy main" > /etc/apt/sources.list.d/neuron.list \
&& curl $([ -n "${NEURON_APT_REPO_KEY}" ] && echo "-u ${NEURON_APT_REPO_KEY}") --retry 3 --retry-delay 1 --retry-all-errors -fSL "https://${NEURON_APT_REPO}/GPG-PUB-KEY-AMAZON-AWS-NEURON.PUB" | gpg --dearmor > /etc/apt/keyrings/neuron.gpg

# Neuron SDK components version numbers
ARG NEURONX_RUNTIME_LIB_VERSION=2.34.10.0-ac18d186d
ARG NEURONX_COLLECTIVES_LIB_VERSION=2.34.10.0-74eaafac6
ARG NEURONX_TOOLS_VERSION=2.32.28.0-526c2b7f6

ARG NEURONX_CC_VERSION=2.27.5334.0+f702b353
ARG NEURONX_JAX_TRAINING_VERSION=0.10.0.1.0.11459+1e1fc1db
ARG NKI_VERSION=0.6.0+31049202112.g85070674
ARG NEURON_AGENTIC_DEVELOPMENT_VERSION=1.3

FROM base AS repo

# Install Neuron components from the apt and pip repos (latest versions)
RUN apt-get update \
&& apt-get install -y \
aws-neuronx-tools \
aws-neuronx-collectives \
aws-neuronx-runtime-lib \
&& rm -rf /var/lib/apt/lists/* \
&& rm -rf /tmp/tmp* \
&& apt-get clean

RUN PIP_REPO_URL=$([ -n "${NEURON_PIP_REPO_KEY}" ] && echo "https://${NEURON_PIP_REPO_KEY}@${NEURON_PIP_REPO}" || echo "https://${NEURON_PIP_REPO}") \
&& ${PIP} install --no-cache-dir --force-reinstall \
--index-url ${PIP_REPO_URL} \
--extra-index-url ${PYPI_SIMPLE_URL} \
--trusted-host ${NEURON_PIP_REPO} \
neuronx-cc \
jax-neuronx \
nki \
neuron_agentic_development \
&& rm -rf ~/.cache/pip/*

FROM base AS prod

# Install Neuron components
# Install Neuron Driver, Runtime and Tools
RUN apt-get update \
&& apt-get install -y \
aws-neuronx-tools=$NEURONX_TOOLS_VERSION \
aws-neuronx-collectives=$NEURONX_COLLECTIVES_LIB_VERSION \
aws-neuronx-runtime-lib=$NEURONX_RUNTIME_LIB_VERSION \
&& rm -rf /var/lib/apt/lists/* \
&& rm -rf /tmp/tmp* \
&& apt-get clean

# Install JAX & Neuron CC
RUN PIP_REPO_URL=$([ -n "${NEURON_PIP_REPO_KEY}" ] && echo "https://${NEURON_PIP_REPO_KEY}@${NEURON_PIP_REPO}" || echo "https://${NEURON_PIP_REPO}") \
&& ${PIP} install --no-cache-dir --force-reinstall \
--index-url ${PIP_REPO_URL} \
--trusted-host ${NEURON_PIP_REPO} \
--extra-index-url ${PYPI_SIMPLE_URL} \
neuronx-cc==$NEURONX_CC_VERSION \
jax-neuronx==$NEURONX_JAX_TRAINING_VERSION \
nki==$NKI_VERSION \
neuron_agentic_development==$NEURON_AGENTIC_DEVELOPMENT_VERSION \
&& rm -rf ~/.cache/pip/*

FROM ${BUILD_STAGE} AS final

# Upgrade OS packages to latest versions
RUN apt-get update \
&& apt-get upgrade -y \
&& rm -rf /var/lib/apt/lists/* \
&& apt-get clean

# Starts framework
ENTRYPOINT ["bash", "-m", "start_with_right_hostname.sh"]
CMD ["/bin/bash"]

HEALTHCHECK CMD curl --fail http://localhost:8080/ping || exit 1
Loading
Loading