MapReduce编程模型

MapReduce/Hadoop

MapReduce是目前云计算中最广泛使用的计算模型,由Google于2004年提出,谷歌关于云计算有三篇著名的论文:

  1. 《Bigtable_A Distributed Storage System for Structured Data》
  2. 《MapReduce: Simplied Data Processing on Large Clusters》
  3. 《The Google File System》

论文下载地址:http://pan.baidu.com/s/1o6G8PGA
Hadoop是MapReduce的一个开源实现,核心框架有2个:HDFS和MapReduce,HDFS为海量数据提供存储,MapReduce为海量数据提供计算。

为什么会有MapReduce?

在MapReduce提出之前,编写并行分布式程序需要下列技术:

  • Multi‐threading( 多线程编程)
  • Socket programming(socket网络编程)
  • Data distribution( 数据分发)
  • Job distribution, coordination, load balancing(任务分发、协调、负载平衡)
  • Fault tolerance( 容错性能)
  • Debugging( 调试)

上面每个方面都需要学习和经验积累,要想编写并行分布式程序并不容易,需要非常有经验的程序员和调试技巧,调试分布式系统很花时间和精力。为了解决这一问题,提出来解决思路:

  • 程序员写串行程序,编程序时不需要思考并行的问题,调试时只需要保证串行执行正确。
  • 由系统完成并行分布式地执行,负责并行分布执行的正确性和效率

但是这样也带来问题:牺牲了程序的功能。直接进行并行分布式编程,可以完成各种各样丰富的功能,而一个编程模型实际上是限定了程序的功能类型。因此要求系统的编程模型必须有代表性,必须代表一大类重要的应用才有生命力。

MapReduce的核心思想是分而治之,把大的任务分成若干个小任务,并行执行小任务,最后把所有的结果汇总。

MapReduce数据模型

MapReduce的数据模型:

  • <key, value>
  • 数据由一条一条的记录组成
  • 记录之间是无序的
  • 每一条记录有一个key,和一个value
  • key: 可以不唯一
  • key与value的具体类型和内部结构由程序员决定,系统基
    本上把它们看作黑匣

图解:

下面以wordcount为例说明MapReduce计算过程:
输入文本:

hello world hadoop hdfs hadoop hello hadoop hdfs

map输出:

<hello,1>
<world,1>
<hadoop,1>
<hdfs,1>
<hadoop,1>
<hello,1>
<hadoop,1>
<hdfs,1>

shuffle(洗牌)过程把key值相同的value合并成list作为reduce输入:

<hello,<1,1>> <world,1> <hadoop,<111>> <hdfs,<1,1>>

reduce输出:

<hello,2>
<world,1>
<hadoop,3>
<hdfs,1>

关于Wordcount运行例子可以参考hadoop helloworld(wordcount),代码解读博客园上有一篇很详细的文章Hadoop集群(第6期)_WordCount运行详解.

附wordcount源码:

import java.io.IOException;
import java.util.StringTokenizer;

import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.io.IntWritable;
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.output.FileOutputFormat;
import org.apache.hadoop.mapred.TextInputFormat;
import org.apache.hadoop.mapred.TextOutputFormat;
import org.apache.hadoop.util.GenericOptionsParser;

public class WordCount {

  // This is the Mapper class
  // reference: http://hadoop.apache.org/docs/r2.6.0/api/org/apache/hadoop/mapreduce/Mapper.html
  //
  public static class TokenizerMapper extends Mapper<Object, Text, Text, IntWritable>{

    private final static IntWritable one = new IntWritable(1);
    private Text word = new Text();

    public void map(Object key, Text value, Context context
                    ) throws IOException, InterruptedException {
      StringTokenizer itr = new StringTokenizer(value.toString());
      while (itr.hasMoreTokens()) {
        word.set(itr.nextToken());
        context.write(word, one);
      }
    }
  }

  public static class IntSumCombiner extends Reducer<Text,IntWritable,Text,IntWritable> {
    private IntWritable result = new IntWritable();

    public void reduce(Text key, Iterable<IntWritable> values,
                       Context context
                       ) throws IOException, InterruptedException {
      int sum = 0;
      for (IntWritable val : values) {
        sum += val.get();
      }
      result.set(sum);
      context.write(key, result);
    }
  }

  // This is the Reducer class
  // reference http://hadoop.apache.org/docs/r2.6.0/api/org/apache/hadoop/mapreduce/Reducer.html
  //
  // We want to control the output format to look at the following:
  //
  // count of word = count
  //
  public static class IntSumReducer extends Reducer<Text,IntWritable,Text,Text> {

    private Text result_key= new Text();
    private Text result_value= new Text();
    private byte[] prefix;
    private byte[] suffix;

    protected void setup(Context context) {
      try {
        prefix= Text.encode("count of ").array();
        suffix= Text.encode(" =").array();
      } catch (Exception e) {
        prefix = suffix = new byte[0];
      }
    }

    public void reduce(Text key, Iterable<IntWritable> values, 
                       Context context
                       ) throws IOException, InterruptedException {
      int sum = 0;
      for (IntWritable val : values) {
        sum += val.get();
      }

      // generate result key
      result_key.set(prefix);
      result_key.append(key.getBytes(), 0, key.getLength());
      result_key.append(suffix, 0, suffix.length);

      // generate result value
      result_value.set(Integer.toString(sum));

      context.write(result_key, result_value);
    }
  }

  public static void main(String[] args) throws Exception {
    Configuration conf = new Configuration();
    String[] otherArgs = new GenericOptionsParser(conf, args).getRemainingArgs();
    if (otherArgs.length < 2) {
      System.err.println("Usage: wordcount <in> [<in>...] <out>");
      System.exit(2);
    }

    Job job = Job.getInstance(conf, "word count");

    job.setJarByClass(WordCount.class);

    job.setMapperClass(TokenizerMapper.class);
    job.setCombinerClass(IntSumCombiner.class);
    job.setReducerClass(IntSumReducer.class);

    job.setMapOutputKeyClass(Text.class);
    job.setMapOutputValueClass(IntWritable.class);

    job.setOutputKeyClass(Text.class);
    job.setOutputValueClass(Text.class);

    // add the input paths as given by command line
    for (int i = 0; i < otherArgs.length - 1; ++i) {
      FileInputFormat.addInputPath(job, new Path(otherArgs[i]));
    }

    // add the output path as given by the command line
    FileOutputFormat.setOutputPath(job,
      new Path(otherArgs[otherArgs.length - 1]));

    System.exit(job.waitForCompletion(true) ? 0 : 1);
  }
}

你可能感兴趣的:(mapreduce,云计算)