spark-env.sh 配置示例

#spark-env.sh 
JAVA_HOME=/home/hadoop/app/jdk1.7.0_60
SCALA_HOME=/home/hadoop/app/scala-2.10.3
SPARK_HOME=/home/hadoop/app/spark-1.4.0
SPARK_PID_DIR=$SPARK_HOME/tmp
HADOOP_CONF_DIR=/home/hadoop/app/hadoop/etc/hadoop
SPARK_CLASSPATH=$SPARK_HOME/conf/:$SPARK_HOME/lib/*:/home/hadoop/app/hadoop/share/hadoop/common/lib/hadoop-lzo-0.4.19.jar:/home/hadoop/app/hbase/conf:/home/hadoop/app/hadoop/lib/native:$SPARK_CLASSPATH
SPARK_JAVA_OPTS="$SPARK_JAVA_OPTS -Dspark.akka.askTimeout=300 -Dspark.ui.retainedStages=1000 -Dspark.eventLog.enabled=true -Dspark.eventLog.dir=hdfs://sparkcluster/user/spark_history_logs -Dspark.shuffle.spill=false -Dspark.shuffle.manager=hash -Dspark.yarn.max.executor.failures=99999 -Dspark.worker.timeout=300"
SPARK_LOCAL_DIRS=/data1/hadoop/spark_local_dir,/data2/hadoop/spark_local_dir,/data3/hadoop/spark_local_dir,/data4/hadoop/spark_local_dir,/data5/hadoop/spark_local_dir,/data6/hadoop/spark_local_dir,/data7/hadoop/spark_local_dir,/data8/hadoop/spark_local_dir,/data9/hadoop/spark_local_dir,/data10/hadoop/spark_local_dir

SPARK_MASTER_PORT=4050
SPARK_WORKER_CORES=30
SPARK_WORKER_MEMORY=60g
SPARK_WORKER_INSTANCES=6
SPARK_DRIVER_MEMORY=12g
SPARK_DAEMON_JAVA_OPTS="-Dspark.deploy.recoveryMode=ZOOKEEPER -Dspark.deploy.zookeeper.url=spark1:2181,spark2:2181,spark3:2181 $SPARK_DAEMON_JAVA_OPTS"

# 总核数 = 物理CPU个数 X 每颗物理CPU的核数 
# 总逻辑CPU数 = 物理CPU个数 X 每颗物理CPU的核数 X 超线程数
## 查看物理CPU个数
cat /proc/cpuinfo| grep "physical id"| sort| uniq| wc -l
# 查看逻辑CPU的个数
cat /proc/cpuinfo| grep "processor"| wc -l