为什么要用使用mapreduce
1、海量数据在单机上处理因为硬件资源限制,无法胜任
2、而一旦将单机版程序扩展到集群来分布式运行,将极大增加程序的复杂度和开发难度
3、引入mapreduce框架后,开发人员可以将绝大部分工作集中在业务逻辑的开发上,而将分布式计算中的复杂性交由框架来处理
4、一些复杂的逻辑使用hive实现起来很麻烦,即使用hive实现了,之后需求变更,改起来也会很麻烦
mapreduce的原理
1、maptask收集我们的map()方法输出的kv对,放到内存缓冲区中
2、从内存缓冲区不断溢出本地磁盘文件,可能会溢出多个文件
3、多个溢出文件会被合并成大的溢出文件
4、在溢出过程中,及合并的过程中,都要调用partitoner进行分组和针对key进行排序
5、reducetask根据自己的分区号,去各个maptask机器上取相应的结果分区数据
6、reducetask会取到同一个分区的来自不同maptask的结果文件,reducetask会将这些文件再进行合并(归并排序)
7、合并成大文件后,shuffle的过程也就结束了,后面进入reducetask的逻辑运算过程(从文件中取出一个一个的键值对group,调用用户自定义的reduce()方法)
mapreduce的shuffle过程
• mapreduce中,map阶段处理的数据如何传递给reduce阶段,是mapreduce框架中最关键的一个流程,这个流程就叫shuffle;
• shuffle: 洗牌、发牌——(核心机制:数据分区,排序,缓存);
• 具体来说:就是将maptask输出的处理结果数据,分发给reducetask,并在分发的过程中,对数据按key进行了分区和排序;
maptask的数据切片
mapreduce代码实例–wordcount
package com.zjs.mapreduce;
import org.apache.commons.io.filefilter.TrueFileFilter;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.FileSystem;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.io.IntWritable;
import org.apache.hadoop.io.LongWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Job;
import org.apache.hadoop.mapreduce.Mapper;
import org.apache.hadoop.mapreduce.Reducer;
import org.apache.hadoop.mapreduce.lib.input.FileInputFormat;
import org.apache.hadoop.mapreduce.lib.input.TextInputFormat;
import org.apache.hadoop.mapreduce.lib.output.FileOutputFormat;
import org.apache.hadoop.mapreduce.lib.output.TextOutputFormat;
import org.apache.log4j.BasicConfigurator;
import java.io.IOException;
/**
* Created by zhangjishuai on 2017/8/6.
*/
public class WordCount {
public static class WordCountMap extends Mapper<LongWritable, Text, Text, IntWritable> {
// @Override
// protected void setup(Context context) throws IOException, InterruptedException {
// super.setup(context);
// }
Text k = new Text();
IntWritable v = new IntWritable();
@Override
protected void map(LongWritable key, Text value, Context context) throws IOException, InterruptedException {
String line = value.toString();
String[] words = line.split(" ");
for (String word : words) {
k.set(word);
v.set(1);
context.write(k, v);
}
}
// @Override
// protected void cleanup(Context context) throws IOException, InterruptedException {
// super.cleanup(context);
// }
}
public static class WordCountReduce extends Reducer<Text, IntWritable, Text, IntWritable> {
// @Override
// protected void setup(Context context) throws IOException, InterruptedException {
// super.setup(context);
// }
IntWritable v = new IntWritable();
@Override
protected void reduce(Text key, Iterable<IntWritable> values, Context context) throws IOException, InterruptedException {
int count = 0;
for (IntWritable v : values) {
count += v.get();
}
v.set(count);
context.write(key, v);
}
// @Override
// protected Object clone() throws CloneNotSupportedException {
// return super.clone();
// }
}
public static void main(String[] args) throws Exception {
Configuration conf = new Configuration();
Job job = Job.getInstance(conf);
BasicConfigurator.configure();
job.setJarByClass(WordCount.class);
job.setMapperClass(WordCountMap.class);
job.setReducerClass(WordCountReduce.class);
job.setMapOutputKeyClass(Text.class);
job.setMapOutputValueClass(IntWritable.class);
job.setOutputKeyClass(Text.class);
job.setOutputValueClass(IntWritable.class);
job.setInputFormatClass(TextInputFormat.class);
job.setOutputFormatClass(TextOutputFormat.class);
FileInputFormat.setInputPaths(job, new Path("/Users/zhangjishuai/zjs/wordcount/input"));
FileSystem fs = FileSystem.get(conf);
Path path = new Path("/Users/zhangjishuai/zjs/wordcount/output");
if(fs.exists(path)){
fs.delete(path,true);
}
FileOutputFormat.setOutputPath(job, path);
job.waitForCompletion(true);
}
}