1.MapReduce做什么
Mapper负责“分”,即把复杂的任务分解为若干个“简单的任务”来处理。
Reducer负责对map阶段的结果进行汇总。
2.MapReduce工作机制
实体一:客户端,用来提交MapReduce作业。
实体二:JobTracker,用来协调作业的运行。
实体三:TaskTracker,用来处理作业划分后的任务。
实体四:HDFS,用来在其它实体间共享作业文件。
3.编写MapReduce程序
1.初始化一个words.txt文件并上传HDFS
2.自定义MapReduce函数
(1)覆盖map函数:继承mapper类并重写方法
public static class MyMapper extends
Mapper<LongWritable, Text, Text, LongWritable> {//Mapper<KEYIN,VALUEIN, KEYOUT, VALUEOUT>
protected void map(LongWritable key, Text value,
Mapper<LongWritable, Text, Text, LongWritable>.Context context)
throws java.io.IOException, InterruptedException {
String[] spilted = value.toString().split(" ");
for (String word : spilted) {
context.write(new Text(word), new LongWritable(1L));
}
};
}
(2)覆盖reduce函数:继承reduce函数并重写其方法
public static class MyReducer extends
Reducer<Text, LongWritable, Text, LongWritable> {//Reducer<KEYIN, VALUEIN, KEYOUT, VALUEOUT>
protected void reduce(Text key,
java.lang.Iterable<LongWritable> values,
Reducer<Text, LongWritable, Text, LongWritable>.Context context)
throws java.io.IOException, InterruptedException {
long count = 0L;
for (LongWritable value : values) {
count += value.get();
}
context.write(key, new LongWritable(count));
};
}
3.设置main函数
(1)指定输入、输出目录
public static final String INPUT_PATH = "hdfs://hadoop-master:9000/testdir/input/words.txt";
public static final String OUTPUT_PATH = "hdfs://hadoop-master:9000/testdir/output/wordcount";
(2)指定自定义的mapper和reducer类
// 1.1:指定对输入数据进行格式化处理的类(可以省略)
job.setInputFormatClass(TextInputFormat.class);
// 1.2:指定自定义的Mapper类
job.setMapperClass(MyMapper.class);
// 1.3:指定map输出的<K,V>类型(如果<k3,v3>的类型与<k2,v2>的类型一致则可以省略)
job.setMapOutputKeyClass(Text.class);
job.setMapOutputValueClass(LongWritable.class);
// 1.4:分区(可以省略)
job.setPartitionerClass(HashPartitioner.class);
// 1.5:设置要运行的Reducer的数量(可以省略)
job.setNumReduceTasks(1);
// 1.6:指定自定义的Reducer类
job.setReducerClass(MyReducer.class);
// 1.7:指定reduce输出的<K,V>类型
job.setOutputKeyClass(Text.class);
job.setOutputValueClass(LongWritable.class);
(3)提交作业
boolean success = job.waitForCompletion(true);
if (success) {
System.out.println("Success");
System.exit(0);
} else {
System.out.println("Failed");
System.exit(1);
}
}
完整代码:
public static void main(String[] args) throws Exception {
Configuration conf = new Configuration();
// 0.0:首先删除输出路径的已有生成文件
FileSystem fs = FileSystem.get(new URI(INPUT_PATH), conf);
Path outPath = new Path(OUTPUT_PATH);
if (fs.exists(outPath)) {
fs.delete(outPath, true);
}
Job job = new Job(conf, "WordCount");
job.setJarByClass(MyWordCountJob.class);
// 1.0:指定输入目录
FileInputFormat.setInputPaths(job, new Path(INPUT_PATH));
// 1.1:指定对输入数据进行格式化处理的类(可以省略)
job.setInputFormatClass(TextInputFormat.class);
// 1.2:指定自定义的Mapper类
job.setMapperClass(MyMapper.class);
// 1.3:指定map输出的<K,V>类型(如果<k3,v3>的类型与<k2,v2>的类型一致则可以省略)
job.setMapOutputKeyClass(Text.class);
job.setMapOutputValueClass(LongWritable.class);
// 1.4:分区(可以省略)
job.setPartitionerClass(HashPartitioner.class);
// 1.5:设置要运行的Reducer的数量(可以省略)
job.setNumReduceTasks(1);
// 1.6:指定自定义的Reducer类
job.setReducerClass(MyReducer.class);
// 1.7:指定reduce输出的<K,V>类型
job.setOutputKeyClass(Text.class);
job.setOutputValueClass(LongWritable.class);
// 1.8:指定输出目录
FileOutputFormat.setOutputPath(job, new Path(OUTPUT_PATH));
// 1.9:指定对输出数据进行格式化处理的类(可以省略)
job.setOutputFormatClass(TextOutputFormat.class);
// 2.0:提交作业
boolean success = job.waitForCompletion(true);
if (success) {
System.out.println("Success");
System.exit(0);
} else {
System.out.println("Failed");
System.exit(1);
}
}