Getting started with hbase

Remarks#

This section provides an overview of what hbase is, and why a developer might want to use it.

It should also mention any large subjects within hbase, and link out to the related topics. Since the Documentation for hbase is new, you may need to create initial versions of those related topics.

Installing HBase in Standalone

HBase Standalone is a mode which allow you to get rid of HDFS and to test HBase before deploying in a cluster, It is not production oriented.

Installing HBase in standalone is extremely simple. First you have to download the HBase archive named hbase-X.X.X-bin.tar.gz available on one of the apache mirrors.

Once you have done this, execute this shell command

tar xzvf hbase-X.X.X-bin.tar.gz

It will export the archive in your directory, you can put it wherever you want.

Now, go to the HBase directory you have exported and edit the file conf/hbase-env.sh

cd hbase-X.X.X
vi -o conf/hbase-env.xml

In this file, uncomment the line and change the path of JAVA_HOME

JAVA_HOME=/usr    #The directory must contain bin/java

Almost there ! now edit the file conf/hbase-sitexml and put the folowing lines

<configuration>
  <property>
    <name>hbase.rootdir</name>
    <value>file:///home/user/hbase</value>
  </property>
  <property>
    <name>hbase.zookeeper.property.dataDir</name>
    <value>/home/user/zookeeper</value>
  </property>
</configuration>

You can put those directories wherever you want to, just be sure to remember it if you want to check logs etc.

Your HBase is now ready to run ! Just execute the command

bin/start-hbase.sh

and if you want to stop HBase

bin/stop-hbase.sh

Now your HBase is launched on your localhost and you can access it (using the Java API or the HBase shell). To run HBase shell, use

bin/hbase shell

Have fun using HBase !

Installing HBase in cluster

TODO