Apache Tajo - Integration with HBase

Apache Tajo supports HBase integration. This enables us to access HBase tables in Tajo. HBase is a distributed column-oriented database built on top of the Hadoop file system. It is a part of the Hadoop ecosystem that provides random real-time read/write access to data in the Hadoop File System. The following steps are required to configure HBase integration.

Set Environment Variable

Add the following changes to “conf/tajo-env.sh” file.

$ vi conf/tajo-env.sh  
# HBase home directory. It is opitional but is required mandatorily to use HBase. 
# export HBASE_HOME = path/to/HBase

After you have included the HBase path, Tajo will set the HBase library file to the classpath.

Create an External Table

Create an external table using the following syntax −

CREATE [EXTERNAL] TABLE [IF NOT EXISTS] <table_name> [(<column_name> <data_type>, ... )] 
USING hbase WITH ('table' = '<hbase_table_name>' 
, 'columns' = ':key,<column_family_name>:<qualifier_name>, ...' 
, 'hbase.zookeeper.quorum' = '<zookeeper_address>' 
, 'hbase.zookeeper.property.clientPort' = '<zookeeper_client_port>') 
[LOCATION 'hbase:zk://<hostname>:<port>/'] ;

To access HBase tables, you must configure the tablespace location.

Here,

  • Table − Set hbase origin table name. If you want to create an external table, the table must exists on HBase.

  • Columns − Key refers to the HBase row key. The number of columns entry need to be equal to the number of Tajo table columns.

  • hbase.zookeeper.quorum − Set zookeeper quorum address.

  • hbase.zookeeper.property.clientPort − Set zookeeper client port.

Query

CREATE EXTERNAL TABLE students (rowkey text,id int,name text) 
USING hbase WITH ('table' = 'students', 'columns' = ':key,info:id,content:name') 
LOCATION 'hbase:zk://<hostname>:<port>/';

Here, the Location path field sets the zookeeper client port id. If you don’t set the port, Tajo will refer the property of hbase-site.xml file.

Create Table in HBase

You can start the HBase interactive shell using the “hbase shell” command as shown in the following query.

Query

/bin/hbase shell

Result

The above query will generate the following result.

hbase(main):001:0>

Steps to Query HBase

To query HBase, you should complete the following steps −

Step 1 − Pipe the following commands to the HBase shell to create a “tutorial” table.

Query

hbase(main):001:0> create ‘students’,{NAME => ’info’},{NAME => ’content’} 
put 'students', ‘row-01', 'content:name', 'Adam' 
put 'students', ‘row-01', 'info:id', '001' 
put 'students', ‘row-02', 'content:name', 'Amit' 
put 'students', ‘row-02', 'info:id', '002' 
put 'students', ‘row-03', 'content:name', 'Bob' 
put 'students', ‘row-03', 'info:id', ‘003'

Step 2 − Now, issue the following command in hbase shell to load the data into a table.

main):001:0> cat ../hbase/hbase-students.txt | bin/hbase shell

Step 3 − Now, return to the Tajo shell and execute the following command to view the metadata of the table −

default> \d students;  

table name: default.students 
table path: 
store type: HBASE 
number of rows: unknown 
volume: 0 B 
Options: 
   'columns' = ':key,info:id,content:name' 
   'table' = 'students'  

schema: 
rowkey  TEXT 
id  INT4 
name TEXT

Step 4 − To fetch the results from the table, use the following query −

Query

default> select * from students

Result

The above query will fetch the following result −

rowkey,  id,  name 
------------------------------- 
row-01,  001,  Adam 
row-02,  002,  Amit 
row-03   003,  Bob
Advertisements