스칼라 설치

sudo apt install scala

sbt 설치

http://www.scala-sbt.org/download.html

spark 다운로드 및 압축 해제

http://d3kbcqa49mib13.cloudfront.net/spark-2.0.0-bin-hadoop2.7.tgz

hadoop 다운로드 및 압축 해제

http://www.apache.org/dyn/closer.cgi/hadoop/common/hadoop-2.7.1/hadoop-2.7.1.tar.gz

.bashrc에 추가

# JAVA_HOME, HADOOP_HOME, SPARK_HOME, 

export JAVA_HOME=C:\ProgramFiles\Java\jdk1.8.0_101 ?
export HADOOP_HOME=C:\utils\hadoop-2.7.1 ??
export SPARK_DIST_CLASSPATH=??  # hadoop classpath
export SPARK_HOME=/usr/local/src/spark-2.0.0-bin-hadoop2.7
export PATH=$SPARK_HOME/bin:$HADOOP_HOME/bin:$PATH

테스트

./bin/example SparkPi 10

https://spark-packages.org/package/mongodb/mongo-spark

다운로드 및 빌드

https://github.com/mongodb/mongo-spark

실행

pyspark --packages org.mongodb.spark:mongo-spark-connector_2.11:1.0.0

.bashrc에 추가

export PYSPARK_DRIVER_PYTHON=ipython
#export PYSPARK_DRIVER_PYTHON_OPTS=qtconsole