spark2.1.0WordCountCluster

1.在本地创建一个文件spark.txt

hello tom
hello jerry
hello kitty
hello world
hello tom
hi nihao
ni ni
hi hi
hello hi
hi hello

2.用winscp 将spark.txt上传到虚拟机上

3.将spark.txt上传到hadoop集群

hadoop  fs  -put spark.txt  /spark,txt

4.在eclipse上写代码如下

package cn.pbj.sparkWcount;

import java.io.Serializable;
import java.util.Iterator;
import java.util.Arrays;
import org.apache.spark.SparkConf;
import org.apache.spark.api.java.*;
import org.apache.spark.api.java.function.*;

import scala.Tuple2;


public class WCCluster implements Serializable {
	    public void execute(String inputPath, String outputFile) {

	       SparkConf conf = new SparkConf().setMaster("local").setAppName("WordCount");
	       JavaSparkContext sc = new JavaSparkContext(conf);

	        JavaRDD input = sc.textFile(inputPath);
	        JavaRDD words = input.flatMap(new FlatMapFunction() {

	            /// https://stackoverflow.com/questions/38880956/spark-2-0-0-arrays-aslist-not-working-incompatible-types/38881118#38881118
	            /// In 2.0, FlatMapFunction.call() returns an Iterator rather than Iterable. Try this:
	            public Iterator call(String x) {
	                return Arrays.asList(x.split(" ")).iterator();
	            }
	        });
	        JavaPairRDD counts = words.mapToPair(new PairFunction() {
	            public Tuple2 call(String x) {
	               return new Tuple2(x, 1);
	            }
	        }).reduceByKey(new Function2() {
	            public Integer call(Integer x, Integer y) throws Exception {
	                return x + y;
	            }
	        });

	        counts.saveAsTextFile(outputFile);
	        sc.stop();
	    }

	    public static void main( String[] args ) {
	        String inputFile = "hdfs://spark01:9000/spark.txt";       //输入路径
	        String outputFile = "/root/wordCount/";     /// 输出路径
	        WCCluster wc = new WCCluster();
	       wc.execute(inputFile, outputFile);

	    }
}

5.将项目文件打包成WcCluster,jar

6.将WcCluster,jar上传到虚拟机中

7.在spark/bin目录下执行

./spark-submit --class cn.pbj.sparkWcount.WCCluster --master spark://spark01:7077 /usr/local/WcCluster.jar

--class 包名.类名  --master主节点  jar路径/jar文件

8.执行后出现INFO output.FileOutputCommitter: Saved output of task 'attempt_20170606235459_0001_m_000000_1' tohdfs://spark01:9000/root/wordCount/_temporary/0/task_20170606235459_0001_m_000000

在spark01:50070、root/wordCount/  输出

下载到本地文件如下

(hi,5)
(tom,2)
(hello,7)
(jerry,1)
(ni,2)
(kitty,1)
(nihao,1)
(world,1)

总结：

1，个人Java基础一般代码只能懂个大概，代码是从GitHub上copy的，还需要提高。

2.就是iterator和iterable两个的区别搞不懂

看别人都是用iterable 但是我的eclipse上就是用不了。

 JavaRDD words = input.flatMap(new FlatMapFunction() {

public Iterator call(String x) { return Arrays.asList(x.split(" ")).iterator(); } });

区别一下

JavaRDD words = lines.flatMap(new FlatMapFunction() {
			
			private static final long serialVersionUID = 1L;
			
			@Override
			public Iterable call(String line) throws Exception {
				return Arrays.asList(line.split(" "));  
			}
			
		});

现在还是搞不懂这两个。