Database Reference
In-Depth Information
Figure 10.6
Choosing a shipping address at checkout
Of course, in addition to a customer's shipping address, other customer
information, such as billing address, preferences, billing credits/debits, and
customer benefits (for example, free shipping) must be stored. For this type of
application, real-time access is required. Thus, the use of the batch processing of
Pig, Hive, or Hadoop's MapReduce is not a reasonable implementation approach.
The following discussion examines how HBase stores the data and provides
real-time read and write access.
As mentioned, HBase is built on top of HDFS. HBase uses a key/value structure
to store the contents of an HBase table. Each value is the data to be stored at the
intersection of the row, column, and version. Each key consists of the following
elements [29]:
• Row length
• Row (sometimes called the row key)
• Column family length
• Column family
• Column qualifier
• Version
• Key type
The
row
is used as the primary attribute to access the contents of an HBase
table. The row is the basis for how the data is distributed across the cluster and
allows a query of an HBase table to quickly retrieve the desired elements. Thus, the
structure or layout of the row has to be specifically designed based on how the data
will be accessed. In this respect, an HBase table is purpose built and is not intended
