HBASE二级索引

HBASE基于coprocessor实现二级索引

场景如下:存储UC_TWEETS表,ROWKEY设计:folderId_dayId_siteId_docId,导出有如下需求:根据campaignId导出,所以需要存储campaignId的索引表

实现步骤如下:

一, 代码实现如下:

public class HbaseCoprocessor extends BaseRegionObserver {

@Override

public void prePut(final ObserverContext e, final Put put,

final WALEdit edit, final Durability durability) throws IOException {

Configuration configuration = HBaseConfiguration.create();

configuration.set("hbase.regionserver.lease.period", "900000");

configuration.set("hbase.rpc.timeout", "1800000");

configuration.set("hbase.client.scanner.timeout.period", "1800000");

configuration.set("hbase.zookeeper.property.clientPort", "2181");

configuration.set("hbase.zookeeper.quorum", "DEV-HADOOP-01,DEV-HADOOP-02,DEV-HADOOP-03");

configuration.set("hbase.master", "DEV-HADOOP-01:60000");

HTable table = new HTable(configuration, "UC_INDEX");

List kv = put.get("f1".getBytes(), "campaignId".getBytes());

Iterator kvItor = kv.iterator();

while (kvItor.hasNext()) {

KeyValue tmp = (KeyValue)kvItor.next();

String rowkey = new String(tmp.getRow());

String value = new String(tmp.getValue());

String newRowkey = value + "_" + rowkey;

Put indexPut = new Put(newRowkey.getBytes());

indexPut.add("f1".getBytes(), tmp.getQualifier(), tmp.getValue());

table.put(indexPut);

}

table.close();

}

}

二, 把上面的HbaseCoprocessor类导出.选择Export -> Jar File,导出成ucTweet.jar文件

三, 把ucTweet.jar文件上传到HDFS,命令如下:./hadoop fs -put /data/server/ucTweet_index.jar /jars

四, 设置UC_TWEETS表的coprocessor属性,命令如下:alter 'UC_TWEETS',METHOD=>'table_att','coprocessor'=>'hdfs://192.168.9.110:9000/jars/ucTweet.jar|com.prime.dsc.inputservice.coprocessor.HbaseCoprocessor|1001|'

五, 把数据插入UC_TWEETS表,如果UC_INDEX表同样有数据,并且符合设计,则说明二级索引建立成功