Spark应用程序开发流程

配置文件:

pom.xml

  <properties>
    <scala.version>2.11.8</scala.version>
    <spark.version>2.2.0</spark.version>
    <hadoop.version>2.6.0-cdh5.7.0</hadoop.version>
  </properties>

  <repositories>
    <!--添加cloudera仓库依赖, CDH版本是cloudera仓库下的-->
    <repository>
      <id>cloudera</id>
      <name>cloudera</name>
      <url>https://repository.cloudera.com/artifactory/cloudera-repos/</url>
    </repository>
  </repositories>
  
    <dependencies>

    <!--添加scala依赖-->
    <dependency>
      <groupId>org.scala-lang</groupId>
      <artifactId>scala-library</artifactId>
      <version>${scala.version}</version>
    </dependency>

    <!--添加spark-code的依赖，scala版本2.11-->
    <dependency>
      <groupId>org.apache.spark</groupId>
      <artifactId>spark-core_2.11</artifactId>
      <version>${spark.version}</version>
    </dependency>

    <!--添加hadoop-client的依赖-->
    <dependency>
      <groupId>org.apache.hadoop</groupId>
      <artifactId>hadoop-client</artifactId>
      <version>${hadoop.version}</version>
    </dependency>

  </dependencies>

---

---

测试代码:

传入参数:


WordCountApp.scala

package com.ruozedata

import org.apache.spark.{SparkConf, SparkContext}

object WordCountApp extends App {

  val conf = new SparkConf()
  val sc = new SparkContext(conf)

  //输入（用args()传入参数,非硬编码）
  val dataFile = sc.textFile(args(0))

  //业务逻辑
  val outputFile = dataFile.flatMap(_.split(",")).map((_,1)).reduceByKey(_+_)

  //输出文件
  outputFile.saveAsTextFile(args(1))

  //关闭流（输入流）
  sc.stop()
}

---

---

CLI中测试:



---

---

打包提交到服务器并执行:










---

---

Linux下本地模式提交到服务器: (在脚本中配置)

$ /home/hadoop/app/spark/bin/spark-submit 
--class com.ruozedata.WordCountApp 
--master local[2] 
--name WordCountApp 
/home/hadoop/lib/spark/SparkCodeApp-1.0.jar 
/wc_input/ /wc_output

具体配置参考Spark官网:

http://spark.apache.org/docs/2.2.0/rdd-programming-guide.html
http://spark.apache.org/docs/2.2.0/configuration.html
http://spark.apache.org/docs/2.2.0/submitting-applications.html