spark操作Kudu之读 - 使用DataFrame API

 

虽然我们可以通过上面显示的KuduContext执行大量操作，但我们还可以直接从默认数据源本身调用读/写API。

要设置读取，我们需要为Kudu表指定选项，命名我们要读取的表以及为表提供服务的Kudu集群的Kudu主服务器列表。

import org.apache.kudu.spark.kudu._
import org.apache.spark.{SparkConf, SparkContext}
import org.apache.spark.sql.SparkSession

/**
  * Created by angel；
  */
object DataFrame_read {
  def main(args: Array[String]): Unit = {
    val sparkConf = new SparkConf().setAppName("AcctfileProcess")
      //设置Master_IP并设置spark参数
      .setMaster("local")
      .set("spark.worker.timeout", "500")
      .set("spark.cores.max", "10")
      .set("spark.rpc.askTimeout", "600s")
      .set("spark.network.timeout", "600s")
      .set("spark.task.maxFailures", "1")
      .set("spark.speculationfalse", "false")
      .set("spark.driver.allowMultipleContexts", "true")
      .set("spark.serializer", "org.apache.spark.serializer.KryoSerializer")
    val sparkContext = SparkContext.getOrCreate(sparkConf)
    val sqlContext = SparkSession.builder().config(sparkConf).getOrCreate().sqlContext
    //TODO 1:定义表名
    val kuduTableName = "spark_kudu_tbl"
    val kuduMasters = "hadoop01:7051,hadoop02:7051,hadoop03:7051"
    //使用spark创建kudu表
    val kuduContext = new KuduContext(kuduTableName, sqlContext.sparkContext)

    //TODO 2：配置kudu参数
    val kuduOptions: Map[String, String] = Map(
      "kudu.table"  -> kuduTableName,
      "kudu.master" -> kuduMasters)
    //TODO 3：执行读取操作
    val customerReadDF = sqlContext.read.options(kuduOptions).kudu
    val filterData = customerReadDF.select("name" ,"age", "city").filter("age<30")
    //TODO 4：打印
    filterData.show()
  }
}