Combiner
1.对combiner的理解
combiner其实属于优化方案,由于带宽限制,应该尽量map和reduce之间的数据传输数量。它在Map端把同一个key的键值对合并在一起并计算,计算规则与reduce一致,所以combiner也可以看作特殊的Reducer。
执行combiner操作要求开发者必须在程序中设置了combiner(程序中通过job.setCombinerClass(myCombine.class)自定义combiner操作)
2.哪里使用combiner?
1,map输出数据根据分区排序完成后,在写入文件之前会执行一次combine操作(前提是作业中设置了这个操作);
2,如果map输出比较大,溢出文件个数大于3(此值可以通过属性min.num.spills.for.combine配置)时,在merge的过程(多个spill文件合并为一个大文件)中前还会执行combiner操作;
3.注意事项
不是每种作业都可以做combiner操作的,只有满足以下条件才可以:
1,combiner只应该用于那种Reduce的输入key/value与输出key/value类型完全一致,因为combine本质上就是reduce操作。
2,计算逻辑上,combine操作后不会影响计算结果,像求和,最大值就不会影响,求平均值就影响了。
4.wordcount的cominer开发实例
package com.sl.bigdatatest.mapreduce;
import java.io.IOException;
import java.net.URI;
import java.net.URISyntaxException;
import java.util.StringTokenizer;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.FileSystem;
import org.apache.hadoop.fs.Path;
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;
public class WordCount {
public static void main(String[] args) throws Exception {
if (args.length < 2) {
System.err.println("Uage:<in> <out>");
System.exit(2);
}
String inputPath = args[0];
Path outputPath = new Path(args[1]);
//1.configuration
Configuration conf = new Configuration();
URI uri = new URI("hdfs://192.168.0.200:9000");
FileSystem fileSystem = FileSystem.get(uri, conf);
if (fileSystem.exists(outputPath)) {
boolean b = fileSystem.delete(outputPath, true);
System.out.println("已存在目录删除:"+b);
}
//2.建立job
Job job = Job.getInstance(conf, WordCount.class.getName());
job.setJarByClass(WordCount.class);
//3.输入文件
FileInputFormat.setInputPaths(job, new Path(inputPath));
//4.格式化输入文件
job.setInputFormatClass(TextInputFormat.class);
//5.map
job.setMapperClass(MapWordCountTask.class);
job.setMapOutputKeyClass(Text.class);
job.setMapOutputValueClass(LongWritable.class);
//6.reduce
job.setReducerClass(ReduceWordCountTask.class);
job.setOutputKeyClass(Text.class);
job.setOutputValueClass(LongWritable.class);
/**指定本job使用combiner组件,组件所用的类为ReduceWordCountTask**/
job.setCombinerClass(ReduceWordCountTask.class);
//7.输出文件
FileOutputFormat.setOutputPath(job, outputPath);
//8.输出文件格式化
job.setOutputFormatClass(TextOutputFormat.class);
//9.提交给集群执行
job.waitForCompletion(true);
}
public static class MapWordCountTask extends Mapper<LongWritable, Text, Text, LongWritable> {
private Text k2 = new Text();
private LongWritable v2 = new LongWritable();
@Override
protected void map(LongWritable key, Text value, Context context) throws Exception {
String content = value.toString();
StringTokenizer st = new StringTokenizer(content);
while (st.hasMoreElements()) {
k2.set(st.nextToken());
v2.set(1L);
context.write(k2, v2);
}
}
}
public static class ReduceWordCountTask extends Reducer<Text, LongWritable, Text, LongWritable> {
private LongWritable v3 = new LongWritable();
@Override
protected void reduce(Text k2, Iterable<LongWritable> v2s,Context context) throws Exception {
long sum = 0;
for (LongWritable longWritable : v2s) {
sum += longWritable.get();
v3.set(sum);
}
context.write(k2, v3);
}
}
}