Difference Hbase, RDBMS Advantages Hbase

Let’s study about the differences and adavntages of HBase.

Hbase
RDBMS
It is column-oriented.It is row-oriented.
Hbase has Flexible schema.RDBMS is fixed schema.
It is good for unstructured and semi-structured data but it also can work with structured data.It is good for structured data only.
HBase is horizontally scalable.RDBMS is hard to scale.
Mainly designed for huge tables.Designed for of small tables.
Suitable for Online Analytical Processing (OLAP).Suitable for Online Transaction Process (OLTP).
It has de-normalized data (process of trying to improve the read performance of a database).It will have normalized data (organized data in the database).
Here Hbase is tightly integrated with MR (MapReduce).RDBMS is not integrated with MR.
Here Join operation in not optimized.It has optimized join operation.
Hive can process Million queries per second.RDBMS can process thousand queries per second.
Here database size varies from TB to PB.Here database size varies from GB to TB.
Supports only for bytes.Supports rich data types.

Advantage of HBase

Hbase provides wide range of benefits and solutions for many use cases. Let’s see one by one,

  1. Hbase can easily handle and store very large volumes of data on top of HDFS file storage. It will aggregate and analyze billions of rows present in the HBase tables.
  2. Hbase data reading and processing takes less time compared to traditional relational models.
  3. Hbase is extensively used in online analytical operations.  Example: In banking applications such as real-time data updates in ATM machines likes getting mini statements,new pin code generation and pin code modification etc.
  4. It has very flexible on schema design.
  5. It supports scaling with Hadoop file system even on commodity hardware.
  6. Hbase is fault tolerance and auto failover in nature.
  7. Hbase is open source software and License free.
  8. This can be integrated with Hive for SQL-like queries, which is helpfull for DBA’s who are more familiar with SQL queries.