Introduction to NoSQL

NoSQL stands for NO SQL that is ,it does not follow traditional SQL with RDBMS instead it is a DBMS with support of large scalability and thus it is emerging as the solution to BigData.

Big data is a popular term used to describe the exponential growth and availability of data, both structured and unstructured.

Difference between NoSQL and SQL?

With increasing amount of data on social networking sites such as posts,tweets,messages etc.Need for fast data processing emerged SQL is strictly RDBMS and has strict Schema but it lacks in Scalability and that is where NoSQL emerges as a hero!

The key differences between them are:

Comparison on	SQL	NoSQL	Winner
Queries	Has good support for all kind of query needs like SELECT,UPDATE,JOINS etc.	Support is currently being added as it lacks the ease with which JOINS are performed in SQL and many other features of a   perfect SQL.	SQL
Speed	Uses query for performing tasks on data.	Uses Map Reduce to perform tasks on data which gives distinct advantage as Map Reduce computes parallel on blocks of a single file.	NoSQL
Transactions	Can commit any number of cross transactions.	Uses Entity Group for Multiple Row Transactions but usage of these Entity Groups for cross transactions is limited to 5 in most cases.	SQL
Scalability	Not scalable to millions of requests at a time.	More scalable than SQL as it uses multiple replication.	NoSQL
Schema	Strictly enforced schema which provides a good hold over how data is being stored and its relationship with others. But for huge table changing schema sometimes locks table temporary this is the main disadvantage.	No locking of table occurs.	NoSQL

To have more grasp on the differences between SQL and NoSQL please watch this video.

Now moving towards the BigData available solutions:-

1. The First Solution is developed and maintained by Google which is known as Data Store.
2. The Second Solution is known as Hadoop and is maintained by Apache.

Data Store

Developed and maintained by Google it is a stack which uses GFS(Google File System) as stroage file system and uses Big Table as NoSQL solution.

It's complete stack consists of :

Hadoop

Hadoop was initially developed at Yahoo and it is a set of tools which uses HDFS in place of GFS of Google and instead of BigTable it uses HBase.

To understand the working of Hadoop more clearly and broad sense.I would recommend you to read the following blog.

http://bradhedlund.com/2011/09/10/understanding-hadoop-clusters-and-the-network/