1.需求：
使用JDBC RDD 操作數(shù)據(jù)庫
2.在數(shù)據(jù)庫中建表并插入數(shù)據(jù)：

create table emp(
                id int(11), 
                ename varchar(20), 
                deptno int(11), 
                sal int(11));

insert into  emp values(1,"Tom",10,2500);
insert into  emp values(2,"Movle",11,1000);
insert into  emp values(2,"Mike",10,1500);
insert into  emp values(2,"jack",11,500);

3.添加JDBC驅(qū)動

4.寫代碼：

MyJDBCRddDemo.scala

import org.apache.spark.SparkConf
import org.apache.spark.SparkContext
import org.apache.spark.rdd.JdbcRDD
import java.sql.DriverManager

/**
 * 使用JDBC RDD 操作數(shù)據(jù)庫
 */


object MyJDBCRddDemo {

  val connection = () => {
    Class.forName("com.mysql.jdbc.Driver").newInstance()
    DriverManager.getConnection("jdbc:mysql://localhost:3306/company?serverTimezone=UTC&characterEncoding=utf-8", "root", "123456")
  }

  def main(args: Array[String]): Unit = {

    //創(chuàng)建Spark對象
    val conf = new SparkConf().setAppName("My JDBC Rdd Demo").setMaster("local")
    val sc = new SparkContext(conf)

    val mysqlRDD = new JdbcRDD(sc,connection,"select * from emp where sal > ? and sal <= ?",900,2000, 2, r=>{
      //獲取員工的姓名和薪水
      val ename = r.getString(2)
      val sal = r.getInt(4)
      (ename,sal)
    })

    val result = mysqlRDD.collect()
    println(result.toBuffer)
    sc.stop
  }
}

5.結(jié)果：