spark-submit提交方式测试Demo

写一个小小的Demo测试一下Spark提交程序的流程

Maven的pom文件

<properties>
        <maven.compiler.source>1.7</maven.compiler.source>
        <maven.compiler.target>1.7</maven.compiler.target>
        <encoding>UTF-8</encoding>
        <spark.version>1.6.1</spark.version>
  </properties>

  <dependencies>
           <dependency>
            <groupId>org.apache.spark</groupId>
            <artifactId>spark-core_2.10</artifactId>
            <version>${spark.version}</version>
        </dependency>

        <dependency>
            <groupId>redis.clients</groupId>
            <artifactId>jedis</artifactId>
            <version>2.7.1</version>
        </dependency>

  </dependencies>
  
   <build>
        <plugins>
            <plugin>
                <groupId>org.apache.maven.plugins</groupId>
                <artifactId>maven-compiler-plugin</artifactId>
                <configuration>
                    <source>1.7</source>
                    <target>1.7</target>
                </configuration>
            </plugin>
        
            <plugin>
                <groupId>org.apache.maven.plugins</groupId>
                <artifactId>maven-shade-plugin</artifactId>
                <version>2.4.3</version>
                <executions>
                    <execution>
                        <phase>package</phase>
                        <goals>
                            <goal>shade</goal>
                        </goals>
                        <configuration>
                            <filters>
                                <filter>
                                    <artifact>*:*</artifact>
                                    <excludes>
                                        <exclude>META-INF/*.SF</exclude>
                                        <exclude>META-INF/*.DSA</exclude>
                                        <exclude>META-INF/*.RSA</exclude>
                                    </excludes>
                                </filter>
                            </filters>
                        </configuration>
                    </execution>
                </executions>
            </plugin>
        </plugins>
    </build>

编写一个蒙特卡罗求PI的代码

import java.util.ArrayList;
import java.util.List;

import org.apache.spark.SparkConf;
import org.apache.spark.api.java.JavaRDD;
import org.apache.spark.api.java.JavaSparkContext;
import org.apache.spark.api.java.function.Function;
import org.apache.spark.api.java.function.Function2;

import redis.clients.jedis.Jedis;

/** 
 * Computes an approximation to pi
 * Usage: JavaSparkPi [slices]
 */
public final class JavaSparkPi {

  public static void main(String[] args) throws Exception {
    SparkConf sparkConf = new SparkConf().setAppName("JavaSparkPi")/*.setMaster("local[2]")*/;
    JavaSparkContext jsc = new JavaSparkContext(sparkConf);
    
    Jedis jedis = new Jedis("192.168.49.151",19000);
    int slices = (args.length == 1) ? Integer.parseInt(args[0]) : 2;
    int n = 100000 * slices;
    List<Integer> l = new ArrayList<Integer>(n);
    for (int i = 0; i < n; i++) {
      l.add(i);
    }

    JavaRDD<Integer> dataSet = jsc.parallelize(l, slices);

    int count = dataSet.map(new Function<Integer, Integer>() {
      @Override
      public Integer call(Integer integer) {
        double x = Math.random() * 2 - 1;
        double y = Math.random() * 2 - 1;
        return (x * x + y * y < 1) ? 1 : 0;
      }
    }).reduce(new Function2<Integer, Integer, Integer>() {
      @Override
      public Integer call(Integer integer, Integer integer2) {
        return integer + integer2;
      }
    });

    jedis.set("Pi", String.valueOf(4.0 * count / n));
    System.out.println("Pi is roughly " + 4.0 * count / n);
    
    jsc.stop();
  }
}

前提条件的setMaster("local[2]") 没有在代码中hard code

本地模式测试情况：# Run application locally on 8 cores

spark-submit
--master local[8]
--class com.spark.test.JavaSparkPi
--executor-memory 4g
--executor-cores 4
/home/dinpay/test/Spark-SubmitTest.jar 100

运行结果在本地：运行在本地一起提交8个Task，不会在WebUI的8080端口上看见提交的任务

-------------------------------------

spark-submit
--master local[8]
--class com.spark.test.JavaSparkPi
--executor-memory 8G
--total-executor-cores 8
hdfs://192.168.46.163:9000/home/test/Spark-SubmitTest.jar 100

运行报错：java.lang.ClassNotFoundException: com.spark.test.JavaSparkPi

------------------------------------

spark-submit
--master local[8]
--deploy-mode cluster
--supervise
--class com.spark.test.JavaSparkPi
--executor-memory 8G
--total-executor-cores 8
/home/dinpay/test/Spark-SubmitTest.jar 100

运行报错:Error: Cluster deploy mode is not compatible with master "local"

====================================================================

Standalone模式client模式 # Run on a Spark standalone cluster in client deploy mode

spark-submit
--master spark://hadoop-namenode-02:7077
--class com.spark.test.JavaSparkPi
--executor-memory 8g
--tital-executor-cores 8
/home/dinpay/test/Spark-SubmitTest.jar 100

运行结果如下：

-------------------------------------------
spark-submit
--master spark://hadoop-namenode-02:7077
--class com.spark.test.JavaSparkPi
--executor-memory 4g
--executor-cores 4g
hdfs://192.168.46.163:9000/home/test/Spark-SubmitTest.jar 100

运行报错:java.lang.ClassNotFoundException: com.spark.test.JavaSparkPi

=======================================================================

standalone模式下的cluster模式 # Run on a Spark standalone cluster in cluster deploy mode with supervise

spark-submit
--master spark://hadoop-namenode-02:7077
--class com.spark.test.JavaSparkPi
--deploy-mode cluster
--supervise
--executor-memory 4g
--executor-cores 4
/home/dinpay/test/Spark-SubmitTest.jar 100

运行报错：java.io.FileNotFoundException: /home/dinpay/test/Spark-SubmitTest.jar (No such file or directory)

-------------------------------------------

spark-submit
--master spark://hadoop-namenode-02:7077
--class com.spark.test.JavaSparkPi
--deploy-mode cluster
--supervise
--driver-memory 4g
--driver-cores 4
--executor-memory 2g
--total-executor-cores 4
hdfs://192.168.46.163:9000/home/test/Spark-SubmitTest.jar 100

运行结果如下：

=============================================

如果代码中写定了.setMaster("local[2]");
则提交的集群模式也会运行driver，但是不会有对应的application并行运行

spark-submit --deploy-mode cluster
--master spark://hadoop-namenode-02:6066
--class com.dinpay.bdp.rcp.service.Window12HzStat
--driver-memory 2g
--driver-cores 2
--executor-memory 1g
--total-executor-cores 2
hdfs://192.168.46.163:9000/home/dinpay/RCP-HZ-TASK-0.0.1-SNAPSHOT.jar
如果代码中限定了.setMaster("local[2]");
则提交方式还是本地模式，会找一台worker进行本地化运行任务