org.apache.hadoop.mapred
Class MapTask.MapOutputBuffer<K,V>

java.lang.Object
  extended by org.apache.hadoop.mapred.MapTask.MapOutputBuffer<K,V>
All Implemented Interfaces:
MapOutputCollector<K,V>, org.apache.hadoop.util.IndexedSortable
Enclosing class:
MapTask

@InterfaceAudience.LimitedPrivate(value="MapReduce")
@InterfaceStability.Unstable
public static class MapTask.MapOutputBuffer<K,V>
extends Object
implements MapOutputCollector<K,V>, org.apache.hadoop.util.IndexedSortable


Nested Class Summary
protected  class MapTask.MapOutputBuffer.BlockingBuffer
          Inner class managing the spill of serialized records to disk.
 class MapTask.MapOutputBuffer.Buffer
           
protected  class MapTask.MapOutputBuffer.InMemValBytes
          Inner class wrapping valuebytes, used for appendRaw.
protected  class MapTask.MapOutputBuffer.MRResultIterator
           
protected  class MapTask.MapOutputBuffer.SpillThread
           
 
Nested classes/interfaces inherited from interface org.apache.hadoop.mapred.MapOutputCollector
MapOutputCollector.Context
 
Constructor Summary
MapTask.MapOutputBuffer()
           
 
Method Summary
 void close()
           
 void collect(K key, V value, int partition)
          Serialize the key, value to intermediate storage.
 int compare(int mi, int mj)
          Compare logical range, st i, j MOD offset capacity.
 void flush()
           
 void init(MapOutputCollector.Context context)
           
 void swap(int mi, int mj)
          Swap logical indices st i, j MOD offset capacity.
 
Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait
 

Constructor Detail

MapTask.MapOutputBuffer

public MapTask.MapOutputBuffer()
Method Detail

init

public void init(MapOutputCollector.Context context)
          throws IOException,
                 ClassNotFoundException
Specified by:
init in interface MapOutputCollector<K,V>
Throws:
IOException
ClassNotFoundException

collect

public void collect(K key,
                    V value,
                    int partition)
             throws IOException
Serialize the key, value to intermediate storage. When this method returns, kvindex must refer to sufficient unused storage to store one METADATA.

Specified by:
collect in interface MapOutputCollector<K,V>
Throws:
IOException

compare

public int compare(int mi,
                   int mj)
Compare logical range, st i, j MOD offset capacity. Compare by partition, then by key.

Specified by:
compare in interface org.apache.hadoop.util.IndexedSortable
See Also:
IndexedSortable.compare(int, int)

swap

public void swap(int mi,
                 int mj)
Swap logical indices st i, j MOD offset capacity.

Specified by:
swap in interface org.apache.hadoop.util.IndexedSortable
See Also:
IndexedSortable.swap(int, int)

flush

public void flush()
           throws IOException,
                  ClassNotFoundException,
                  InterruptedException
Specified by:
flush in interface MapOutputCollector<K,V>
Throws:
IOException
ClassNotFoundException
InterruptedException

close

public void close()
Specified by:
close in interface MapOutputCollector<K,V>


Copyright © 2013 Apache Software Foundation. All Rights Reserved.