Accessing HBase data with hive

The Hive architecture is by design scalable and can easily scale for hundreds or millions of users; clubbing it with Hbase makes the right integration choice. It also make Hbase interoperate with Hive; a simple Hbase table can be accessed from Hive as if it was a Hive table itself.

A single hive query can perform joins, unions, and Cartesian joins by aggregating a combination of Hbase and Hive tables.

Similarly, the inserts can also be applied between Hbase and native Hive tables.

In a nutshell, we can solve the following usecase:

• Incremental refresh problems by keeping a near real-time replica of MySql data in Hbase

• We can pick and chose when to use Hbase and when to use native Hive tables as hive tables are much faster to read and write.

How to do it …

Connecting Hive to Hbase:

1. I am assuming your EMR cluster is still running and you are able to connect using putty or directly ssh to the cluster.

2. Write hive in the command shell and you will be connected to the hive...