How is NO SQL different?

Till date we have all worked with RDBMS systems which support SQL as a query language and the same can be used to retrieve any desired amount and combination of data from the RDBMS systems. Whenever the data volume of an enterprise grows, the company needs to scale up its RDBMS infrastructure by adding more servers to store the increased volume of data.

However, going by the rate at which enterprise data volume grows it is becoming more an more evident that structured storage mediums would no longer be able to handle the huge volume of data to be generated in future by sources such as social media.

NOSQL collectively refers to a class of database management system which stores the data in an unstructured format. There you would not have fixed table structures and joins and hence do not involve any SQL. These database systems store data in an unstructured format distributed across multiple storage locations.These database systems can hence scale horizontally.

The NOSQL databases follow a distributed architecture and can service heavy read/write workloads.
Some of the advantages of NOSQL database management systems are:
  • Easy and cheap horizontal scaling
  • Can handle very huge volume of data
  • Flexible data models can be followed.

Some of the disadvantages are:
  • Not enough matured concept for critical business case implementations
  • Reliable support still an issue as compared to the RDBMS systems provided by large corporations
  • Redundant and and unreliable data
  • Unstructured data which cannot act as a reliable source for high end BI applications
The well known Apache Hadoop is a NOSQL unstructured database management system, also an open source one. Google File System is another one of the same kind developed by Google.