Skip to content
This repository has been archived by the owner on Feb 3, 2021. It is now read-only.

Feature: nvBLAS and OpenBLAS plugin #539

Merged
merged 10 commits into from
May 16, 2018
Merged

Feature: nvBLAS and OpenBLAS plugin #539

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
10 commits
Select commit Hold shift + click to select a range
0a242ee
add openblas plugin, update gpu docker images with netlib-lgpl
jafreck May 3, 2018
1990c55
Merge branch 'master' into feature/spark-openblas-plugin
jafreck May 3, 2018
40ce7ec
update images and plugins
jafreck May 4, 2018
cf97a04
Merge branch 'feature/spark-openblas-plugin' of github.com:jafreck/az…
jafreck May 4, 2018
e94c125
Merge branch 'master' into feature/spark-openblas-plugin
jafreck May 10, 2018
8938190
add nvblas plugin
jafreck May 10, 2018
85b2a69
Merge branch 'feature/spark-openblas-plugin' of github.com:jafreck/az…
jafreck May 10, 2018
1678c06
revert gpu docker image change, add -Pnetlib-lgpl to base images
jafreck May 11, 2018
934e84e
change configuraitons to functions, add pugins to cluster.yaml
jafreck May 16, 2018
e35e443
Merge branch 'master' into feature/spark-openblas-plugin
jafreck May 16, 2018
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
1 change: 1 addition & 0 deletions aztk/models/plugins/internal/plugin_manager.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -23,6 +23,7 @@ class PluginManager:
hdfs=plugins.HDFSPlugin,
simple=plugins.SimplePlugin,
spark_ui_proxy=plugins.SparkUIProxyPlugin,
openblas=plugins.OpenBLASPlugin,
)

def __init__(self):
Expand Down
1 change: 1 addition & 0 deletions aztk/spark/models/plugins/__init__.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -5,3 +5,4 @@
from .rstudio_server import RStudioServerPlugin
from .simple import SimplePlugin
from .spark_ui_proxy import SparkUIProxyPlugin
from .openblas import OpenBLASPlugin
1 change: 1 addition & 0 deletions aztk/spark/models/plugins/openblas/__init__.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
from .configuration import *
18 changes: 18 additions & 0 deletions aztk/spark/models/plugins/openblas/configuration.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,18 @@
import os
from aztk.models.plugins.plugin_configuration import PluginConfiguration, PluginPort, PluginTargetRole
from aztk.models.plugins.plugin_file import PluginFile
from aztk.utils import constants

dir_path = os.path.dirname(os.path.realpath(__file__))

class OpenBLASPlugin(PluginConfiguration):
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Same

def __init__(self):
super().__init__(
name="openblas",
ports=[],
target_role=PluginTargetRole.All,
execute="openblas.sh",
files=[
PluginFile("openblas.sh", os.path.join(dir_path, "openblas.sh")),
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Would it make more sense to have a generic apt-get install plugin?

The way I did this for the conda package installer was to let users specify an array of packages as a parameter. The main issue in this case though is the update-alternatives command which is non-standard... not sure how to get around that.

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Update alternatives is borderline not necessary since we are running in a container with a known environment and I already know that there aren't any alternatives present. It's mostly there as a precaution. I think that a general apt-get install plugin would be great, but I don't know if it should be in place of this.

],
)
4 changes: 4 additions & 0 deletions aztk/spark/models/plugins/openblas/openblas.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,4 @@
#!/bin/bash
apt-get update
apt-get install -y libopenblas-base
update-alternatives --config libblas.so.3
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Does it make sense to just bake this into the dockerfile?

157 changes: 156 additions & 1 deletion docker-image/gpu/spark1.6.3/Dockerfile
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,4 +1,159 @@
FROM aztk/spark:v0.1.0-spark1.6.3-base
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

why not build on top of this base image any more?

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

There is a difference between base and gpu images in the Spark building process now (the added -Pnetlib-lgpl). We could build all images with that profile -- that might be the best option here.

# Ubuntu 16.04 (Xenial)
FROM ubuntu:16.04

# set AZTK version compatibility
ENV AZTK_DOCKER_IMAGE_VERSION 0.1.0

# set version of python required for aztk
ENV AZTK_PYTHON_VERSION=3.5.2

# modify these ARGs on build time to specify your desired versions of Spark/Hadoop
ENV SPARK_VERSION_KEY 1.6.3
ENV SPARK_FULL_VERSION spark-${SPARK_VERSION_KEY}-bin-without-hadoop
ENV HADOOP_VERSION 2.8.3
ENV LANG=C.UTF-8 LC_ALL=C.UTF-8

# set env vars
ENV JAVA_HOME /usr/lib/jvm/java-1.8.0-openjdk-amd64
ENV SPARK_HOME /home/spark-current
ENV PATH $SPARK_HOME/bin:$PATH

RUN apt-get clean \
&& apt-get update -y \
# install dependency packages
&& apt-get install -y --no-install-recommends \
make \
build-essential \
zlib1g-dev \
libssl-dev \
libbz2-dev \
libreadline-dev \
libsqlite3-dev \
maven \
wget \
curl \
llvm \
git \
libncurses5-dev \
libncursesw5-dev \
python3-pip \
python3-venv \
xz-utils \
tk-dev \
&& apt-get update -y \
# install [software-properties-common]
# so we can use [apt-add-repository] to add the repository [ppa:webupd8team/java]
# from which we install Java8
&& apt-get install -y --no-install-recommends software-properties-common \
&& apt-add-repository ppa:webupd8team/java -y \
&& apt-get update -y \
# install java
&& apt-get install -y --no-install-recommends default-jdk \
# set up user python and aztk python
&& ln -s /usr/bin/python3.5 /usr/bin/python \
&& /usr/bin/python -m pip install --upgrade pip setuptools wheel \
&& apt-get remove -y python3-pip \
# build and install spark
&& git clone https://github.com/apache/spark.git \
&& cd spark \
&& git checkout tags/v${SPARK_VERSION_KEY} \
&& export MAVEN_OPTS="-Xmx3g -XX:ReservedCodeCacheSize=1024m" \
&& ./make-distribution.sh --name custom-spark --tgz -Pnetlib-lgpl -Phive -Phive-thriftserver -Dhadoop.version=${HADOOP_VERSION} -Phadoop-2.6 -DskipTests \
&& tar -xvzf /spark/spark-${SPARK_VERSION_KEY}-bin-custom-spark.tgz --directory=/home \
&& ln -s "/home/spark-${SPARK_VERSION_KEY}-bin-custom-spark" /home/spark-current \
&& rm -rf /spark \
# copy azure storage jars and dependencies to $SPARK_HOME/jars
&& echo "<project>" \
"<modelVersion>4.0.0</modelVersion>" \
"<groupId>groupId</groupId>" \
"<artifactId>artifactId</artifactId>" \
"<version>1.0</version>" \
"<dependencies>" \
"<dependency>" \
"<groupId>org.apache.hadoop</groupId>" \
"<artifactId>hadoop-azure-datalake</artifactId>" \
"<version>${HADOOP_VERSION}</version>" \
"<exclusions>" \
"<exclusion>" \
"<groupId>org.apache.hadoop</groupId>" \
"<artifactId>hadoop-common</artifactId>" \
"</exclusion>" \
"</exclusions> " \
"</dependency>" \
"<dependency>" \
"<groupId>org.apache.hadoop</groupId>" \
"<artifactId>hadoop-azure</artifactId>" \
"<version>${HADOOP_VERSION}</version>" \
"<exclusions>" \
"<exclusion>" \
"<groupId>org.apache.hadoop</groupId>" \
"<artifactId>hadoop-common</artifactId>" \
"</exclusion>" \
"<exclusion>" \
"<groupId>com.fasterxml.jackson.core</groupId>" \
"<artifactId>jackson-core</artifactId>" \
"</exclusion>" \
"</exclusions> " \
"</dependency>" \
"<dependency>" \
"<groupId>com.microsoft.sqlserver</groupId>" \
"<artifactId>mssql-jdbc</artifactId>" \
"<version>6.4.0.jre8</version>" \
"</dependency>" \
"<dependency>" \
"<groupId>com.microsoft.azure</groupId>" \
"<artifactId>azure-storage</artifactId>" \
"<version>2.2.0</version>" \
"<exclusions>" \
"<exclusion>" \
"<groupId>com.fasterxml.jackson.core</groupId>" \
"<artifactId>jackson-core</artifactId>" \
"</exclusion>" \
"<exclusion>" \
"<groupId>org.apache.commons</groupId>" \
"<artifactId>commons-lang3</artifactId>" \
"</exclusion>" \
"<exclusion>" \
"<groupId>org.slf4j</groupId>" \
"<artifactId>slf4j-api</artifactId>" \
"</exclusion>" \
"</exclusions>" \
"</dependency>" \
"<dependency>" \
"<groupId>com.microsoft.azure</groupId>" \
"<artifactId>azure-cosmosdb-spark_2.1.0_2.11</artifactId>" \
"<version>1.1.1</version>" \
"<exclusions>" \
"<exclusion>" \
"<groupId>org.apache.tinkerpop</groupId>" \
"<artifactId>tinkergraph-gremlin</artifactId>" \
"</exclusion>" \
"<exclusion>" \
"<groupId>org.apache.tinkerpop</groupId>" \
"<artifactId>spark-gremlin</artifactId>" \
"</exclusion>" \
"<exclusion>" \
"<groupId>io.netty</groupId>" \
"<artifactId>*</artifactId>" \
"</exclusion>" \
"<exclusion>" \
"<groupId>com.fasterxml.jackson.core</groupId>" \
"<artifactId>jackson-annotations</artifactId>" \
"</exclusion>" \
"</exclusions> " \
"</dependency>" \
"</dependencies>" \
"</project>" > /tmp/pom.xml \
&& cd /tmp \
&& mvn dependency:copy-dependencies -DoutputDirectory="${SPARK_HOME}/jars/" \
# cleanup
&& apt-get --purge autoremove -y maven python3-pip \
&& apt-get autoremove -y \
&& apt-get autoclean -y \
&& rm -rf /tmp/* \
&& rm -rf /root/.cache \
&& rm -rf /root/.m2 \
&& rm -rf /var/lib/apt/lists/*

LABEL com.nvidia.volumes.needed="nvidia_driver"

Expand Down
157 changes: 156 additions & 1 deletion docker-image/gpu/spark2.1.0/Dockerfile
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,4 +1,159 @@
FROM aztk/spark:v0.1.0-spark2.1.0-base
# Ubuntu 16.04 (Xenial)
FROM ubuntu:16.04

# set AZTK version compatibility
ENV AZTK_DOCKER_IMAGE_VERSION 0.1.0

# set version of python required for aztk
ENV AZTK_PYTHON_VERSION=3.5.2

# modify these ARGs on build time to specify your desired versions of Spark/Hadoop
ENV SPARK_VERSION_KEY 2.1.0
ENV SPARK_FULL_VERSION spark-${SPARK_VERSION_KEY}-bin-without-hadoop
ENV HADOOP_VERSION 2.8.3
ENV LANG=C.UTF-8 LC_ALL=C.UTF-8

# set env vars
ENV JAVA_HOME /usr/lib/jvm/java-1.8.0-openjdk-amd64
ENV SPARK_HOME /home/spark-current
ENV PATH $SPARK_HOME/bin:$PATH

RUN apt-get clean \
&& apt-get update -y \
# install dependency packages
&& apt-get install -y --no-install-recommends \
make \
build-essential \
zlib1g-dev \
libssl-dev \
libbz2-dev \
libreadline-dev \
libsqlite3-dev \
maven \
wget \
curl \
llvm \
git \
libncurses5-dev \
libncursesw5-dev \
python3-pip \
python3-venv \
xz-utils \
tk-dev \
&& apt-get update -y \
# install [software-properties-common]
# so we can use [apt-add-repository] to add the repository [ppa:webupd8team/java]
# from which we install Java8
&& apt-get install -y --no-install-recommends software-properties-common \
&& apt-add-repository ppa:webupd8team/java -y \
&& apt-get update -y \
# install java
&& apt-get install -y --no-install-recommends default-jdk \
# set up user python and aztk python
&& ln -s /usr/bin/python3.5 /usr/bin/python \
&& /usr/bin/python -m pip install --upgrade pip setuptools wheel \
&& apt-get remove -y python3-pip \
# build and install spark
&& git clone https://github.com/apache/spark.git \
&& cd spark \
&& git checkout tags/v${SPARK_VERSION_KEY} \
&& export MAVEN_OPTS="-Xmx3g -XX:ReservedCodeCacheSize=1024m" \
&& ./dev/make-distribution.sh --name custom-spark --pip --tgz -Pnetlib-lgpl -Phive -Phive-thriftserver -Dhadoop.version=${HADOOP_VERSION} -DskipTests \
&& tar -xvzf /spark/spark-${SPARK_VERSION_KEY}-bin-custom-spark.tgz --directory=/home \
&& ln -s "/home/spark-${SPARK_VERSION_KEY}-bin-custom-spark" /home/spark-current \
&& rm -rf /spark \
# copy azure storage jars and dependencies to $SPARK_HOME/jars
&& echo "<project>" \
"<modelVersion>4.0.0</modelVersion>" \
"<groupId>groupId</groupId>" \
"<artifactId>artifactId</artifactId>" \
"<version>1.0</version>" \
"<dependencies>" \
"<dependency>" \
"<groupId>org.apache.hadoop</groupId>" \
"<artifactId>hadoop-azure-datalake</artifactId>" \
"<version>${HADOOP_VERSION}</version>" \
"<exclusions>" \
"<exclusion>" \
"<groupId>org.apache.hadoop</groupId>" \
"<artifactId>hadoop-common</artifactId>" \
"</exclusion>" \
"</exclusions> " \
"</dependency>" \
"<dependency>" \
"<groupId>org.apache.hadoop</groupId>" \
"<artifactId>hadoop-azure</artifactId>" \
"<version>${HADOOP_VERSION}</version>" \
"<exclusions>" \
"<exclusion>" \
"<groupId>org.apache.hadoop</groupId>" \
"<artifactId>hadoop-common</artifactId>" \
"</exclusion>" \
"<exclusion>" \
"<groupId>com.fasterxml.jackson.core</groupId>" \
"<artifactId>jackson-core</artifactId>" \
"</exclusion>" \
"</exclusions> " \
"</dependency>" \
"<dependency>" \
"<groupId>com.microsoft.sqlserver</groupId>" \
"<artifactId>mssql-jdbc</artifactId>" \
"<version>6.4.0.jre8</version>" \
"</dependency>" \
"<dependency>" \
"<groupId>com.microsoft.azure</groupId>" \
"<artifactId>azure-storage</artifactId>" \
"<version>2.2.0</version>" \
"<exclusions>" \
"<exclusion>" \
"<groupId>com.fasterxml.jackson.core</groupId>" \
"<artifactId>jackson-core</artifactId>" \
"</exclusion>" \
"<exclusion>" \
"<groupId>org.apache.commons</groupId>" \
"<artifactId>commons-lang3</artifactId>" \
"</exclusion>" \
"<exclusion>" \
"<groupId>org.slf4j</groupId>" \
"<artifactId>slf4j-api</artifactId>" \
"</exclusion>" \
"</exclusions>" \
"</dependency>" \
"<dependency>" \
"<groupId>com.microsoft.azure</groupId>" \
"<artifactId>azure-cosmosdb-spark_${SPARK_VERSION_KEY}_2.11</artifactId>" \
"<version>1.1.1</version>" \
"<exclusions>" \
"<exclusion>" \
"<groupId>org.apache.tinkerpop</groupId>" \
"<artifactId>tinkergraph-gremlin</artifactId>" \
"</exclusion>" \
"<exclusion>" \
"<groupId>org.apache.tinkerpop</groupId>" \
"<artifactId>spark-gremlin</artifactId>" \
"</exclusion>" \
"<exclusion>" \
"<groupId>io.netty</groupId>" \
"<artifactId>*</artifactId>" \
"</exclusion>" \
"<exclusion>" \
"<groupId>com.fasterxml.jackson.core</groupId>" \
"<artifactId>jackson-annotations</artifactId>" \
"</exclusion>" \
"</exclusions> " \
"</dependency>" \
"</dependencies>" \
"</project>" > /tmp/pom.xml \
&& cd /tmp \
&& mvn dependency:copy-dependencies -DoutputDirectory="${SPARK_HOME}/jars/" \
# cleanup
&& apt-get --purge autoremove -y maven python3-pip \
&& apt-get autoremove -y \
&& apt-get autoclean -y \
&& rm -rf /tmp/* \
&& rm -rf /root/.cache \
&& rm -rf /root/.m2 \
&& rm -rf /var/lib/apt/lists/*

LABEL com.nvidia.volumes.needed="nvidia_driver"

Expand Down
Loading