安装SPARK和SCALA

1、下载 spark

创新互联建站专注为客户提供全方位的互联网综合服务，包含不限于成都网站设计、成都网站制作、驿城网络推广、成都微信小程序、驿城网络营销、驿城企业策划、驿城品牌公关、搜索引擎seo、人物专访、企业宣传片、企业代运营等，从售前售中售后，我们都将竭诚为您服务，您的肯定，是我们最大的嘉奖；创新互联建站为所有大学生创业者提供驿城建站搭建服务，24小时服务热线：18980820575，官方网址：www.cdcxhl.com

http://mirrors.cnnic.cn/apache/spark/spark-1.3.0/spark-1.3.0-bin-hadoop2.3.tgz

2、下载scala

http://www.scala-lang.org/download/2.10.5.html

3、安装scala

mkdir /usr/lib/scala

tar –zxvf scala-2.10.5.tgz

mv scala-2.10.5 /usr/lib/scala

4、设置scala路径

vim /etc/bashrc

export SCALA_HOME=/usr/lib/scala/scala-2.10.5

export PATH=$SCALA_HOME/bin:$PATH

source /etc/bashrc

scala –version

5、分发

scp -r /usr/lib/scala/ hd2:/usr/lib/scala

scp -r /usr/lib/scala/ hd3:/usr/lib/scala

scp -r /usr/lib/scala/ hd4:/usr/lib/scala

scp -r /usr/lib/scala/ hd5:/usr/lib/scala

scp /etc/bashrc hd2:/etc/bashrc

scp /etc/bashrc hd3:/etc/bashrc

scp /etc/bashrc hd4:/etc/bashrc

scp /etc/bashrc hd5:/etc/bashrc

6、安装spark

tar -zxvf spark-1.3.0-bin-hadoop2.3.tgz

mkdir /usr/local/spark

mv spark-1.3.0-bin-hadoop2.3 /usr/local/spark

vim /etc/bashrc

export SPARK_HOME=/usr/local/spark/spark-1.3.0-bin-hadoop2.3

export PATH=$SCALA_HOME/bin:$SPARK_HOME/bin:$PATH

source /etc/bashrc

cd /usr/local/spark/spark-1.3.0-bin-hadoop2.3/conf/

cp spark-env.sh.template spark-env.sh

vim spark-env.sh

export JAVA_HOME=/java

export SCALA_HOME=/usr/lib/scala/scala-2.10.5

export SPARK_HOME=/usr/local/spark/spark-1.3.0-bin-hadoop2.3

export SPARK_MASTER_IP=192.168.137.101

export SPARK_WORKER_MEMORY=10g

export SPARK_DRIVER_MEMORY=9g

export HADOOP_CONF_DIR=/home/hadoop/hadoop/etc/hadoop

export SPARK_LIBRARY_PATH=$SPARK_HOME/lib

export SCALA_LIBRARY_PATH=$SPARK_LIBRARY_PATH

cp slaves.template slaves

vim slaves

hd1

hd2

hd3

hd4

hd5

7、分发

scp /etc/bashrc hd2:/etc

scp /etc/bashrc hd3:/etc

scp /etc/bashrc hd4:/etc

scp /etc/bashrc hd5:/etc

scp -r /usr/local/spark/spark-1.3.0-bin-hadoop2.3 hd2:/usr/local/spark/

scp -r /usr/local/spark/spark-1.3.0-bin-hadoop2.3 hd3:/usr/local/spark/

scp -r /usr/local/spark/spark-1.3.0-bin-hadoop2.3 hd4:/usr/local/spark/

scp -r /usr/local/spark/spark-1.3.0-bin-hadoop2.3 hd5:/usr/local/spark/

7、启动

在hd1，启动

cd $SPARK_HOME/sbin

./start-all.sh