1)默认partition分区
public class HashPartitioner<K, V> extends Partitioner<K, V> {
/** Use {@link Object#hashCode()} to partition. */
public int getPartition(K key, V value, int numReduceTasks) {
return (key.hashCode() & Integer.MAX_VALUE) % numReduceTasks;
}
}
默认分区是根据key的hashCode对reduceTasks个数取模得到的。用户没法控制哪个key存储到哪个分区
2)自定义Partitioner步骤
(1)自定义类继承Partitioner,重新getPartition()方法
package com.atguigu.mapreduce.flow;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Partitioner;
public class FlowPartitioner extends Partitioner<Text,FlowBean>{
@Override
public int getPartition(Text key, FlowBean value, int numPartitions) {
//1 需求:根据电话号码的前三位是几来分区
//拿到电话号码的前三位
String phoneNum = key.toString().substring(0, 3);
int partitions = 4;
if ("135".equals(phoneNum)) {
partitions = 0;
}else if("136".equals(phoneNum)) {
partitions = 1;
}else if ("137".equals(phoneNum)) {
partitions = 2;
}else if ("138".equals(phoneNum)) {
partitions = 3;
}
return partitions;
}
}
(2)在job驱动中,设置自定义partitioner
job.setPartitionerClass(CustomPartitioner.class)
(3)自定义partition后,要根据自定义partitioner的逻辑设置相应数量的reduce task
job.setNumReduceTasks(5);
3)注意:
如果reduceTask的数量> getPartition的结果数,则会多产生几个空的输出文件part-r-000xx;
如果1<reduceTask的数量<getPartition的结果数,则有一部分分区数据无处安放,会Exception;
如果reduceTask的数量=1,则不管mapTask端输出多少个分区文件,最终结果都交给这一个reduceTask,最终也就只会产生一个结果文件 part-r-00000;
例如:假设自定义分区数为5,则
(1)job.setNumReduceTasks(1);会正常运行,只不过会产生一个输出文件
(2)job.setNumReduceTasks(2);会报错
(3)job.setNumReduceTasks(6);大于5,程序会正常运行,会产生空文件