使用maven构建spark程序，借助IDEA。

README.md

使用maven构建spark程序，借助IDEA。

1. 建立一个Maven工程，项目模版可以选择 maven-archetype-quickstart ；
2. 在项目处右键 Add Framework Support > scala，然后编辑 pom.xml 添加相应依赖；
3. 在main目录下添加一个scala目录，用于放置scala代码，如下SubmitExample.scala；
4. IDEA内右键Run... 可以测试整个工程是否运行正常；
5. 真实提交任务到YARN集群，需要去掉.setMaster("local")代码，并用spark-submit提交任务。

pom.xml

<?xml version="1.0" encoding="UTF-8"?>
<project xmlns="http://maven.apache.org/POM/4.0.0"
         xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
         xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
    <modelVersion>4.0.0</modelVersion>
 
    <groupId>net.jqian.learning.spark</groupId>
    <artifactId>introduction</artifactId>
    <version>1.0-SNAPSHOT</version>
 
    <dependencies>
        <dependency>
            <groupId>org.apache.spark</groupId>
            <artifactId>spark-core_2.10</artifactId>
            <version>1.6.0</version>
        </dependency>
    </dependencies>
 
    <build>
        <plugins>
            <plugin>
                <!-- compile & package scala program -->
                <groupId>net.alchim31.maven</groupId>
                <artifactId>scala-maven-plugin</artifactId>
                <version>3.1.0</version>
                <executions>
                    <execution>
                        <id>compile-scala</id>
                        <phase>compile</phase>
                        <goals>
                            <goal>add-source</goal>
                            <goal>compile</goal>
                        </goals>
                    </execution>
                    <execution>
                        <id>test-compile-scala</id>
                        <phase>test-compile</phase>
                        <goals>
                            <goal>add-source</goal>
                            <goal>testCompile</goal>
                        </goals>
                    </execution>
                </executions>
            </plugin>
            <plugin>
                <!-- package all upstream dependencies -->
                <artifactId>maven-assembly-plugin</artifactId>
                <version>2.5</version>
                <configuration>
                    <descriptorRefs>
                        <descriptorRef>jar-with-dependencies</descriptorRef>
                    </descriptorRefs>
                </configuration>
                <executions>
                    <execution>
                        <id>make-assembly</id>
                        <!-- this is used for inheritance merges -->
                        <phase>package</phase>
                        <!-- bind to the packaging phase -->
                        <goals>
                            <goal>single</goal>
                        </goals>
                    </execution>
                </executions>
            </plugin>
        </plugins>
    </build>
</project>

run.sh

#!/bin/bash

spark-submit \
        --class net.jqian.learning.spark.SubmitExample \
        --num-executors 5 \
        --driver-memory 4g \
        --executor-memory 8g \
        --executor-cores 4 \
        --master yarn-cluster \
        --conf spark.yarn.executor.memoryOverhead=1024 \
        example-0.1-jar-with-dependencies.jar 

SubmitExample.scala

package net.jqian.learning.spark
 
import org.apache.spark.{SparkConf, SparkContext}
import org.apache.spark.SparkContext._

object SubmitExample {
  def main (args: Array[String]) {
    val conf = new SparkConf()
      .setAppName("submitExample")
      .setMaster("local") // 用于本地IDEA测试
    val sc = new SparkContext(conf)
 
    val br = sc.broadcast(Map(1 -> "h", 2 -> "e", 3 -> "l", 4 -> "l", 5 -> "o"))
    val ac = sc.accumulator(0)
    val arr = sc.parallelize(1 to 8 toList).map(
      k => {
        br.value.get(k) match {
          case Some(v) => v
          case None => {
            ac add 1
            ""
          }
        }
      }
    )
    arr.cache()
    val res = arr.reduce(_ + _)
    println(res)
    println("miss found: " + ac.value)
 
    sc.stop()
  }
}