de.hpi.fgis.voidgen.hadoop.tasks.distinctclustering
Class LeftPartPartitioner

java.lang.Object
  extended by org.apache.hadoop.mapreduce.Partitioner<K,V>
      extended by org.apache.hadoop.mapreduce.lib.partition.HashPartitioner<StringIntPair,StringIntPair>
          extended by de.hpi.fgis.voidgen.hadoop.tasks.distinctclustering.LeftPartPartitioner

public class LeftPartPartitioner
extends org.apache.hadoop.mapreduce.lib.partition.HashPartitioner<StringIntPair,StringIntPair>

Partitions the data only dependent on the real-key part of the whole key.

Author:
Martin Linkhorst, Hasso Plattner Institute at University of Potsdam, Germany, Johannes Gosda, Hasso Plattner Institute at University of Potsdam, Germany

Constructor Summary
LeftPartPartitioner()
           
 
Method Summary
 int getPartition(StringIntPair key, StringIntPair value, int numReduceTasks)
           
 
Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait
 

Constructor Detail

LeftPartPartitioner

public LeftPartPartitioner()
Method Detail

getPartition

public int getPartition(StringIntPair key,
                        StringIntPair value,
                        int numReduceTasks)
Overrides:
getPartition in class org.apache.hadoop.mapreduce.lib.partition.HashPartitioner<StringIntPair,StringIntPair>