Hadoop2.2.0+Hive0.13+Hbase0.96.2集成

本篇，散仙主要讲的是使用Hive如何和Hbase集成，Hbase和Hive的底层存储都在HDFS上，都是hadoop生态系统中的重要一员，所以他们之间有着很亲密的联系，可以相互转换与操作。


hadoop，hbase和hive的搭建就不重复说了，不会的朋友，可以看散仙前面的博客，下面直接进入重点，关于hive集成hbase这一块，网上资料不算多，有的版本比较旧，散仙这里使用的版本是hive0.12和hbase0.96.2。


hive集成hbase的方式其实很简单，就是把Hbase的几个jar包，拷贝到hive的lib目录下即可截图如下：




上面的几个包放进Hive的lib中之后，我们需要配置在hive-site.xml里面配置下Hbase的zk地址，便于Hive连接Hbase使用，添加配置如下：

<pre name="code" class="xml">

<property>
<name>hbase.zookeeper.quorum</name>
<value>h1,h2,h3</value>
</property>
</pre>
下面可以进行测试，我们直接把Hbase里的一张表，转成我们的Hive表，Hive启动前，首先要把Hadoop和Hbase启动起来，然后启动Hive，启动完进程如下所示：

<pre name="code" class="java">[search@h1 ~]$ jps
7950 RunJar
6559 DataNode
7691 HRegionServer
6879 ResourceManager
7563 HMaster
6985 NodeManager
7474 HQuorumPeer
6731 SecondaryNameNode
6458 NameNode
8120 Jps
[search@h1 ~]$
</pre>
先在hbase里建一张表，并添加列簇，和列数据如下：



<pre name="code" class="java">hbase(main):001:0> list
TABLE                                                                                                                                                                          
SLF4J: Class path contains multiple SLF4J bindings.
SLF4J: Found binding in [jar:file:/home/search/hbase-0.96.2-hadoop2/lib/slf4j-log4j12-1.6.4.jar!/org/slf4j/impl/StaticLoggerBinder.class]
SLF4J: Found binding in [jar:file:/home/search/hadoop/share/hadoop/common/lib/slf4j-log4j12-1.7.5.jar!/org/slf4j/impl/StaticLoggerBinder.class]
SLF4J: See http://www.slf4j.org/codes.html#multiple_bindings for an explanation.
mytest                                                                                                                                                                         
webpage                                                                                                                                                                        
2 row(s) in 3.0870 seconds

=> ["mytest", "webpage"]
hbase(main):002:0> scan 'mytest'
ROW                                           COLUMN+CELL                                                                                                                      
2207                                         column=item:name, timestamp=1407329588864, value=hadoop                                                                          
23008                                        column=home:address, timestamp=1407329589330, value=shenzhen                                                                     
2309                                         column=info:age, timestamp=1407329589497, value=899                                                                              
24008                                        column=home:address, timestamp=1407329588587, value=guangzhou                                                                    
2407                                         column=item:name, timestamp=1407329588447, value=big data                                                                        
2409                                         column=info:age, timestamp=1407329588728, value=22.32                                                                            
3109                                         column=info:age, timestamp=1407329588308, value=7.4                                                                              
321008                                       column=home:address, timestamp=1407329588169, value=luoyang                                                                      
33107                                        column=item:name, timestamp=1407329588028, value=x-code                                                                          
4205                                         column=home:address, timestamp=1407329587701, value=beijing                                                                      
44101                                        column=info:age, timestamp=1407329587883, value=18.5                                                                             
444105                                       column=item:name, timestamp=1407329587555, value=a dog                                                                           
12 row(s) in 0.0980 seconds
</pre>

在Hive里，执行如下语句，进行关联建表：

<pre name="code" class="java">hive> CREATE EXTERNAL  TABLE hbaseive(key int, name string,address string,age double)          
    > STORED BY 'org.apache.hadoop.hive.hbase.HBaseStorageHandler'    
    > WITH SERDEPROPERTIES ("hbase.columns.mapping" = ":key,item:name,home:address,info:age")
    > TBLPROPERTIES ("hbase.table.name" = "mytest"); </pre>
如果不报错，就证明，集成成功，如果出现错误，一般都是hbase的某个jar包缺少造成，我们根据提示添加到hive的lib里即可，然后重启hive的的shell客户端。


建表成功后，查询如下：


<pre name="code" class="java">hive> select * from hbaseive;
OK
2207    hadoop  NULL    NULL
23008   NULL    shenzhen        NULL
2309    NULL    NULL    899.0
24008   NULL    guangzhou       NULL
2407    big data        NULL    NULL
2409    NULL    NULL    22.32
3109    NULL    NULL    7.4
321008  NULL    luoyang NULL
33107   x-code  NULL    NULL
4205    NULL    beijing NULL
44101   NULL    NULL    18.5
444105  a dog   NULL    NULL
Time taken: 0.481 seconds, Fetched: 12 row(s)
hive> </pre>
至此，我们的Hive集成Hbase就成功了，关联成功后，我们就可以对数据进行分析了，需要注意的是，我们关联过的hbase表，如果在Hbase里面删除了这个表，那么在Hive里面再次查询，就会报错，如果删除了Hive里面的表，则对Hbase没有影响。另外一点需要注意的是，使用Hive关联Hbase的建的表，是一个内部表，所以需要在Create 后面加上 external关键字 。




参考资料：https://cwiki.apache.org/confluence/display/Hive/HBaseIntegration