From f6ca89967fd63d6f3d38d7904bb33d6b98f7545c Mon Sep 17 00:00:00 2001
From: workingloong <WangQL1201@outlook.com>
Date: Wed, 12 Aug 2020 11:41:33 +0800
Subject: [PATCH 1/3] Install Horovod and dependencies in the dev image

---
 elasticdl/docker/Dockerfile | 31 ++++++++++++++++++++++++++++---
 1 file changed, 28 insertions(+), 3 deletions(-)

diff --git a/elasticdl/docker/Dockerfile b/elasticdl/docker/Dockerfile
index 91da12b6e..f9a344c4f 100644
--- a/elasticdl/docker/Dockerfile
+++ b/elasticdl/docker/Dockerfile
@@ -7,9 +7,25 @@ ARG EXTRA_PYPI_INDEX=https://pypi.org/simple
 COPY elasticdl/docker/bashrc /etc/bash.bashrc
 RUN chmod a+rx /etc/bash.bashrc
 
-RUN apt-get -qq update && \
-        apt-get -qq install -y unzip curl git software-properties-common g++ wget \
-        shellcheck libeigen3-dev clang-format > /dev/null && \
+RUN apt-get -qq update && apt-get -qq install -y \
+        unzip \
+        curl \
+        git \
+        software-properties-common \
+        g++ \
+        wget \
+        build-essential \
+        cmake \
+        vim \
+        ca-certificates \
+        libjpeg-dev \
+        libpng-dev \
+        librdmacm1 \
+        libibverbs1 \
+        ibverbs-providers \
+        shellcheck \
+        libeigen3-dev \
+        clang-format > /dev/null && \
         python -m pip install --quiet --upgrade pip
 
 COPY elasticdl_client/requirements.txt /requirements.txt
@@ -61,6 +77,8 @@ COPY elasticdl/python/data/recordio_gen/heart_recordio_gen.py /scripts/heart_rec
 
 FROM dev as allreduce
 
+RUN pip install future typing
+
 # Note that pip is having issue downloading PyTorch on manylinux so we use curl
 # to download it instead
 RUN curl -sLo torch-1.4.0-cp36-cp36m-manylinux1_x86_64.whl \
@@ -72,3 +90,10 @@ RUN cd /root && git clone --depth=1 https://github.com/caicloud/ftlib.git
 RUN cd /root/ftlib && python -m pip install --quiet -r requirements.txt
 RUN cd /root/ftlib/ftlib/consensus/gossip && bash ./gen_shared_lib.sh
 RUN cp -r /root/ftlib/ftlib /usr/local/lib/python3.6/dist-packages/ftlib
+
+ENV HOROVOD_PATH /tmp/horovod
+RUN cd /tmp \
+        && git clone --recursive https://github.com/horovod/horovod.git
+
+RUN cd ${HOROVOD_PATH} && HOROVOD_WITH_TENSORFLOW=1 HOROVOD_WITH_PYTORCH=1  \
+    python setup.py install
\ No newline at end of file

From f54f63de43a04d2e8d7e5767193f27b985034f1b Mon Sep 17 00:00:00 2001
From: workingloong <WangQL1201@outlook.com>
Date: Wed, 12 Aug 2020 11:50:13 +0800
Subject: [PATCH 2/3] Add a blank line at the end

---
 elasticdl/docker/Dockerfile | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/elasticdl/docker/Dockerfile b/elasticdl/docker/Dockerfile
index f9a344c4f..dcbdbe1de 100644
--- a/elasticdl/docker/Dockerfile
+++ b/elasticdl/docker/Dockerfile
@@ -96,4 +96,4 @@ RUN cd /tmp \
         && git clone --recursive https://github.com/horovod/horovod.git
 
 RUN cd ${HOROVOD_PATH} && HOROVOD_WITH_TENSORFLOW=1 HOROVOD_WITH_PYTORCH=1  \
-    python setup.py install
\ No newline at end of file
+    python setup.py install

From 1927a6b4b6db57dd5c6e77f96e0c31b462db1d3b Mon Sep 17 00:00:00 2001
From: workingloong <WangQL1201@outlook.com>
Date: Wed, 12 Aug 2020 16:02:02 +0800
Subject: [PATCH 3/3] Set commit id to clone

---
 elasticdl/docker/Dockerfile | 17 ++++++++++++-----
 1 file changed, 12 insertions(+), 5 deletions(-)

diff --git a/elasticdl/docker/Dockerfile b/elasticdl/docker/Dockerfile
index dcbdbe1de..2d22605c4 100644
--- a/elasticdl/docker/Dockerfile
+++ b/elasticdl/docker/Dockerfile
@@ -3,6 +3,7 @@ ARG BASE_IMAGE
 
 FROM ${BASE_IMAGE} as dev
 ARG EXTRA_PYPI_INDEX=https://pypi.org/simple
+ARG HOROVOD_COMMIT_ID="3108a24"
 
 COPY elasticdl/docker/bashrc /etc/bash.bashrc
 RUN chmod a+rx /etc/bash.bashrc
@@ -91,9 +92,15 @@ RUN cd /root/ftlib && python -m pip install --quiet -r requirements.txt
 RUN cd /root/ftlib/ftlib/consensus/gossip && bash ./gen_shared_lib.sh
 RUN cp -r /root/ftlib/ftlib /usr/local/lib/python3.6/dist-packages/ftlib
 
-ENV HOROVOD_PATH /tmp/horovod
+# The latest package of Horovod does not support elastic training,
+# so we need to git clone and install it using source codes.
+ENV HOROVOD_PATH /tmp/${HOROVOD_COMMIT_ID}
 RUN cd /tmp \
-        && git clone --recursive https://github.com/horovod/horovod.git
-
-RUN cd ${HOROVOD_PATH} && HOROVOD_WITH_TENSORFLOW=1 HOROVOD_WITH_PYTORCH=1  \
-    python setup.py install
+        && git clone --recursive https://github.com/horovod/horovod.git \
+        -b master ${HOROVOD_COMMIT_ID}
+
+RUN cd ${HOROVOD_PATH} && HOROVOD_WITHOUT_MPI=1 \
+        HOROVOD_WITHOUT_MXNET=1 \
+        HOROVOD_WITH_TENSORFLOW=1 \
+        HOROVOD_WITH_PYTORCH=1  \
+        python setup.py install