spark系列三：sparkstreaming基于hdfs实时wordcount案例

HDFS文件
基于HDFS文件的实时计算，其实就是，监控一个HDFS目录，只要其中有新文件出现，就实时处理。相当于处理实时的文件流。

streamingContext.fileStream<KeyClass, ValueClass, InputFormatClass>(dataDirectory)
streamingContext.fileStream[KeyClass, ValueClass, InputFormatClass](dataDirectory)

Spark Streaming会监视指定的HDFS目录，并且处理出现在目录中的文件。要注意的是，所有放入HDFS目录中的文件，都必须有相同的格式；必须使用移动或者重命名的方式，将文件移入目录；一旦处理之后，文件的内容即使改变，也不会再处理了；基于HDFS文件的数据源是没有Receiver的，因此不会占用一个cpu core。
java版本：

package cn.spark.study.core;

import java.util.Arrays;

import org.apache.spark.SparkConf;
import org.apache.spark.api.java.function.FlatMapFunction;
import org.apache.spark.api.java.function.Function2;
import org.apache.spark.api.java.function.PairFunction;
import org.apache.spark.streaming.Durations;
import org.apache.spark.streaming.api.java.JavaDStream;
import org.apache.spark.streaming.api.java.JavaPairDStream;
import org.apache.spark.streaming.api.java.JavaStreamingContext;

import scala.Tuple2;

public class hdfswordcount {
 public static void main(String[] args) {
  SparkConf conf = new SparkConf()
    .setAppName("streamingwordcount")
    .setMaster("local[2]");
  JavaStreamingContext jsc = new JavaStreamingContext(conf,Durations.seconds(1));
  JavaDStream<String> lines = jsc.textFileStream("hdfs://master:9000/streamwordcount");
 
  JavaDStream<String> words = lines.flatMap(new FlatMapFunction<String,String>(){

   private static final long serialVersionUID = 1L;

   @Override
   public Iterable<String> call(String line) throws Exception {
    // TODO Auto-generated method stub
    
    return Arrays.asList(line.split(" "));
   }
   
  });
  JavaPairDStream<String,Integer> pairs = words.mapToPair(new PairFunction<String,String,Integer>(){

   private static final long serialVersionUID = 1L;

   @Override
   public Tuple2<String, Integer> call(String word) throws Exception {
    // TODO Auto-generated method stub
    return new Tuple2<String,Integer>(word,1);
   }
   
  });
  JavaPairDStream<String,Integer> res = pairs.reduceByKey(new Function2<Integer,Integer,Integer>(){

   
   private static final long serialVersionUID = 1L;

   @Override
   public Integer call(Integer v1, Integer v2) throws Exception {
    // TODO Auto-generated method stub
    return v1 + v2;
   }
   
  });
  res.print();
  jsc.start();
  jsc.awaitTermination();
 }

}

scala版本：

package com.spark.study.core

import org.apache.spark.streaming.StreamingContext
import org.apache.spark.streaming.Durations
import org.apache.spark.SparkConf

object hdfswordcount {
  def main(args:Array[String]){
    val conf = new SparkConf()
                .setAppName("hdfswordcount")
                .setMaster("local[2]")
    val scc = new StreamingContext(conf,Durations.seconds(1))
    val lines = scc.textFileStream("hdfs://master:9000/streamwordcount")
    val words = lines.flatMap(x => x.split(" "))
    val pairs = words.map(x =>(x,1))
    val res = pairs.reduceByKey(_ + _ )
    Thread.sleep(5000)
    res.print()
    scc.start()
    scc.awaitTermination()
   
  }
}

来自 “ ITPUB博客 ” ，链接：http://blog.itpub.net/30541278/viewspace-2155524/，如需转载，请注明出处，否则将追究法律责任。

转载于:http://blog.itpub.net/30541278/viewspace-2155524/