HBase components and Know what......

It is one of the cool projects from Apache, that enable a facility to provide a large scale scale able, distributed database, based on Hadoop. In this data is organised as row columns that can grow infinitely as you add up new nodes. No need to reconfigure and mess up much with the configuration setting.

This requires Java and Hadoop to run full fledged manner.

Components:

HBaseMaster :

The HBaseMaster is responsible for assigning regions to HRegionServers. The first region to be assigned is the ROOT region which locates all the META regions to be assigned. The HBaseMaster also monitors the health of each HRegionServer, and if it detects a HRegionServer is no longer reachable, it will split the HRegionServer's write-ahead log so that there is now one write-ahead log for each region that the HRegionServer was serving. After it has accomplished this, it will reassign the regions that were being served by the unreachable HRegionServer. In addition, the HBaseMaster is also responsible for handling table administrative functions such as on/off-lining of tables, changes to the table schema (adding and removing column families), etc.

HRegionServer:

The HRegionServer is responsible for handling client read and write requests. It communicates with the BaseMaster to get a list of regions to serve and to tell the master that it is alive. Region assignments and

other instructions from the master "piggy back" on the heart beat messages.

HBase client:

The HBase client is responsible for finding HRegionServers that are serving the particular row range of interest. On instantiation, the HBase client communicates with the HBaseMaster to find the location of the ROOT region. This is the only communication between the client and the master.

Inherited from  : Here 

From : Research paper of (Ankur Khetrapal, Vinay Ganesh)