Difference between SQL and NoSQL database

Key Difference

	SQL	NoSQL
Known As	SQL Based database Management System is majorly known as RDBMS or DBMS Systems	Whereas NoSQL database are primarily non-relational database or distributed database
Architecture	SQL Database mainly   architecture surrounded to the Tables	In Case of NoSQL database there are major four categorization Document base  Key-value pair base   Graph database  Wide-column Store
Schema	SQL database has its own predefine schema to store structure data	In NoSQL database, there is no predefine schema, here schema is most dynamic element based on the data elements
Scalability	SQL Databases are vertically scalable, means if we want to scale SQL base database, we need to give hardware boost on which the DBMS System is installed. This is where it sometimes goes for the limitation of scalability.	NoSQL database are horizontally scalable, means if we want to scale it, we need to add more nodes and create distribution network based on our own need and required power. This is how it reduce load on the database
Data Retrieval	In SQL based database, to define and manipulate data we can use SQL (Structure Query Language), which is very powerful now a days	In terms of NoSQL database queries are focus on the collection and documents. Sometimes it is called as UnQL (Unstructured Query Language). This is still in the evolution phase, so it is varies from vendor to vendor of the NoSQL database
Systems in Market Place	Oracle MySql Postgres MS-SQL	MongoDB BigTable Redis RavenDb Cassendra Hbase Neo4j CouchDb
Classification	We can classify RDBMS into two major type  Open Source database Close Source database	We can classify NoSQL DB based on the way of storing data Key-value pair store database   Graph database   Document Store   Column Store   XML Store
Based Concept	All RDBMS System based on the codes law and ACID Properties	NoSQL database mostly follow “The CAP Theorem”1

Best Fit for

Complex Query Execution	SQL database are best fit for complex queries intensive environment and relative data	NoSQL don’t have standard interface to perform complex queries and the queries themselves not so powerful in NoSQL database
Type of data store	SQL database are best fit for storing Relational data or sometimes Object oriented data	NoSQL database is best fit for hierarchical data and document base data. This type of database is preferred for large data sets
Transaction based application fitness	SQL databases are excellent for heavy duty transaction based data or related application, as it is more stable and mostly satisfied ACID Properties (Atomicity, Consistency, Isolation and Durability).	NoSQL database basically not meant for transaction based application but it’s main objective focus on document based large data sets
Support	Excellent support available for most of the RDBMS vendors	For some NoSQL database we still need to rely on community support

Advantages of SQL Database

• SQL base RDBMS are excellent in terms of providing atomicity, consistency, isolation and durability
• It can very easily handle large volume of transaction with high level of accuracy.

Advantages of NoSQL Database

• NoSQL database generally process data faster. Reason why whatever data store in the NoSQL database dynamic schema is in the form of the directly consumable JSON Object.
• Simple data model to store data in database
• Major systems are flexible enough to better enable developers to use the application in ways that meet their needs.

For more on NoSQL Database have a look at Brian's NoSQL presentation. It's funny!

Conclusion

Both the data management techniques are best in what they do. Criticizing any one of them will not help. NoSQL databases are great invention over a period of time in order to store large volume of data along with easy and smooth retrieval.    

The CAP Theorem

Impossible for any shared data system to guarantee simultaneously all of the following three properties:

         Consistency: Once data is written all future read requests will contain data

         Availability: The database is always available and responsive

         Partition Tolerance: if part of the database is unavailable, other parts are unaffected.