For use of HDFS and Hadoop, please see my other post.

Setting up HBase as standalone is comparatively simpler.  The only trouble you would face is the Version Mismatching exception between the server and client in which case you would have ideally used a different jar on classpath for the client program and a different version of the HBase server would be running on.

There is only one thing that you need to do to configure HBase as standalone.

a)  Let JAVA_HOME variable in your hbase-env.sh (bin directory) point to the right path.

You can then play with the hbase shell utility

(from the bin folder) > hbase shell

Type in ”help” to get a list of options available.

You can also use the example program available in http://hadoop.apache.org/hbase/docs/r0.18.0/api/index.html to use Java API to talk with HBase.

For use of the example program, you will be need to create a table on HBase. Use hbase shell to execute the following DDL statement.

create ‘myTable’,’myColumnFamily’

machine/ (HBase and Hadoop)