ml.shifu.guagua.mapreduce
类 GuaguaMapper<MASTER_RESULT extends ml.shifu.guagua.io.Bytable,WORKER_RESULT extends ml.shifu.guagua.io.Bytable>

java.lang.Object
  org.apache.hadoop.mapreduce.Mapper<org.apache.hadoop.io.LongWritable,org.apache.hadoop.io.Text,org.apache.hadoop.io.Text,org.apache.hadoop.io.Text>
      ml.shifu.guagua.mapreduce.GuaguaMapper<MASTER_RESULT,WORKER_RESULT>

public class GuaguaMapper<MASTER_RESULT extends ml.shifu.guagua.io.Bytable,WORKER_RESULT extends ml.shifu.guagua.io.Bytable>
extends org.apache.hadoop.mapreduce.Mapper<org.apache.hadoop.io.LongWritable,org.apache.hadoop.io.Text,org.apache.hadoop.io.Text,org.apache.hadoop.io.Text>

GuaguaMapper is the Hadoop Mapper implementation for both guagua master and guagua workers.

Use (GuaguaInputSplit) context.getInputSplit() to check whether this task is guagua master or guagua worker.

guaguaService is the interface for both guagua Master and Worker implementation. According to isMaster, master service and worker service will be determined.

Only mapper, no reducer for guagua MapReduce implementation. And in this mapper run(org.apache.hadoop.mapreduce.Mapper.Context) is override while Mapper.map(Object, Object, org.apache.hadoop.mapreduce.Mapper.Context) is not since we don't need to iterate mapper raw input.

嵌套类摘要

从类 org.apache.hadoop.mapreduce.Mapper 继承的嵌套类/接口

org.apache.hadoop.mapreduce.Mapper.Context

构造方法摘要

GuaguaMapper()

方法摘要

protected void	cleanup(org.apache.hadoop.mapreduce.Mapper.Context context)
ml.shifu.guagua.GuaguaService	getGuaguaService()
boolean	isMaster()
void	run(org.apache.hadoop.mapreduce.Mapper.Context context) Run guagua service according isMaster setting.
void	setGuaguaService(ml.shifu.guagua.GuaguaService guaguaService)
void	setMaster(boolean isMaster)
protected void	setup(org.apache.hadoop.mapreduce.Mapper.Context context)

从类 org.apache.hadoop.mapreduce.Mapper 继承的方法

map

从类 java.lang.Object 继承的方法

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

构造方法详细信息

GuaguaMapper

public GuaguaMapper()

方法详细信息

setup

protected void setup(org.apache.hadoop.mapreduce.Mapper.Context context)
              throws IOException,
                     InterruptedException

覆盖：

类 org.apache.hadoop.mapreduce.Mapper<org.apache.hadoop.io.LongWritable,org.apache.hadoop.io.Text,org.apache.hadoop.io.Text,org.apache.hadoop.io.Text> 中的 setup

抛出：

IOException

InterruptedException

run

public void run(org.apache.hadoop.mapreduce.Mapper.Context context)
         throws IOException,
                InterruptedException

Run guagua service according isMaster setting. Iteration, coordination will be included in service running.

cleanup(org.apache.hadoop.mapreduce.Mapper.Context) is called in finally block to make sure resources can be cleaned.

Guagua try best to update progress for each iteration. And also task status will be updated in each iteration in hadoop job web ui.

覆盖：

类 org.apache.hadoop.mapreduce.Mapper<org.apache.hadoop.io.LongWritable,org.apache.hadoop.io.Text,org.apache.hadoop.io.Text,org.apache.hadoop.io.Text> 中的 run

抛出：

IOException

InterruptedException

cleanup

protected void cleanup(org.apache.hadoop.mapreduce.Mapper.Context context)
                throws IOException,
                       InterruptedException

覆盖：

类 org.apache.hadoop.mapreduce.Mapper<org.apache.hadoop.io.LongWritable,org.apache.hadoop.io.Text,org.apache.hadoop.io.Text,org.apache.hadoop.io.Text> 中的 cleanup

抛出：

IOException

InterruptedException

isMaster

public boolean isMaster()

setMaster

public void setMaster(boolean isMaster)

getGuaguaService

public ml.shifu.guagua.GuaguaService getGuaguaService()

setGuaguaService

public void setGuaguaService(ml.shifu.guagua.GuaguaService guaguaService)