View Javadoc

1   /**
2    *
3    * Licensed to the Apache Software Foundation (ASF) under one
4    * or more contributor license agreements.  See the NOTICE file
5    * distributed with this work for additional information
6    * regarding copyright ownership.  The ASF licenses this file
7    * to you under the Apache License, Version 2.0 (the
8    * "License"); you may not use this file except in compliance
9    * with the License.  You may obtain a copy of the License at
10   *
11   *     http://www.apache.org/licenses/LICENSE-2.0
12   *
13   * Unless required by applicable law or agreed to in writing, software
14   * distributed under the License is distributed on an "AS IS" BASIS,
15   * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
16   * See the License for the specific language governing permissions and
17   * limitations under the License.
18   */
19  package org.apache.hadoop.hbase.mapreduce;
20  
21  import java.io.IOException;
22  
23  import org.apache.commons.logging.Log;
24  import org.apache.commons.logging.LogFactory;
25  import org.apache.hadoop.classification.InterfaceAudience;
26  import org.apache.hadoop.classification.InterfaceStability;
27  import org.apache.hadoop.conf.Configurable;
28  import org.apache.hadoop.conf.Configuration;
29  import org.apache.hadoop.hbase.HBaseConfiguration;
30  import org.apache.hadoop.hbase.client.HTable;
31  import org.apache.hadoop.hbase.io.ImmutableBytesWritable;
32  import org.apache.hadoop.hbase.util.Bytes;
33  import org.apache.hadoop.mapreduce.Partitioner;
34  
35  /**
36   * This is used to partition the output keys into groups of keys.
37   * Keys are grouped according to the regions that currently exist
38   * so that each reducer fills a single region so load is distributed.
39   *
40   * <p>This class is not suitable as partitioner creating hfiles
41   * for incremental bulk loads as region spread will likely change between time of
42   * hfile creation and load time. See {@link LoadIncrementalHFiles}
43   * and <a href="http://hbase.apache.org/docs/current/bulk-loads.html">Bulk Load</a>.
44   *
45   * @param <KEY>  The type of the key.
46   * @param <VALUE>  The type of the value.
47   */
48  @InterfaceAudience.Public
49  @InterfaceStability.Stable
50  public class HRegionPartitioner<KEY, VALUE>
51  extends Partitioner<ImmutableBytesWritable, VALUE>
52  implements Configurable {
53  
54    private static final Log LOG = LogFactory.getLog(HRegionPartitioner.class);
55    private Configuration conf = null;
56    private HTable table;
57    private byte[][] startKeys;
58  
59    /**
60     * Gets the partition number for a given key (hence record) given the total
61     * number of partitions i.e. number of reduce-tasks for the job.
62     *
63     * <p>Typically a hash function on a all or a subset of the key.</p>
64     *
65     * @param key  The key to be partitioned.
66     * @param value  The entry value.
67     * @param numPartitions  The total number of partitions.
68     * @return The partition number for the <code>key</code>.
69     * @see org.apache.hadoop.mapreduce.Partitioner#getPartition(
70     *   java.lang.Object, java.lang.Object, int)
71     */
72    @Override
73    public int getPartition(ImmutableBytesWritable key,
74        VALUE value, int numPartitions) {
75      byte[] region = null;
76      // Only one region return 0
77      if (this.startKeys.length == 1){
78        return 0;
79      }
80      try {
81        // Not sure if this is cached after a split so we could have problems
82        // here if a region splits while mapping
83        region = table.getRegionLocation(key.get()).getRegionInfo().getStartKey();
84      } catch (IOException e) {
85        LOG.error(e);
86      }
87      for (int i = 0; i < this.startKeys.length; i++){
88        if (Bytes.compareTo(region, this.startKeys[i]) == 0 ){
89          if (i >= numPartitions-1){
90            // cover if we have less reduces then regions.
91            return (Integer.toString(i).hashCode()
92                & Integer.MAX_VALUE) % numPartitions;
93          }
94          return i;
95        }
96      }
97      // if above fails to find start key that match we need to return something
98      return 0;
99    }
100 
101   /**
102    * Returns the current configuration.
103    *
104    * @return The current configuration.
105    * @see org.apache.hadoop.conf.Configurable#getConf()
106    */
107   @Override
108   public Configuration getConf() {
109     return conf;
110   }
111 
112   /**
113    * Sets the configuration. This is used to determine the start keys for the
114    * given table.
115    *
116    * @param configuration  The configuration to set.
117    * @see org.apache.hadoop.conf.Configurable#setConf(
118    *   org.apache.hadoop.conf.Configuration)
119    */
120   @Override
121   public void setConf(Configuration configuration) {
122     this.conf = HBaseConfiguration.create(configuration);
123     try {
124       this.table = new HTable(this.conf,
125         configuration.get(TableOutputFormat.OUTPUT_TABLE));
126     } catch (IOException e) {
127       LOG.error(e);
128     }
129     try {
130       this.startKeys = this.table.getStartKeys();
131     } catch (IOException e) {
132       LOG.error(e);
133     }
134   }
135 }