Novel approach for transformation of data from MySQL to NoSQL (MongoDB)

Abstract

Relational databases have been the dominant model for the several decades, for storing and retrieving data. But in some cases relational database has shown its age and losing its importance due to its dependence on fixed schema and inability to scale well. With the development of the internet and cloud services like Google and Amazon volume of data is increasing exponentially. Data accumulated by social services such as Facebook and Twitter is more connected and semi-structured. Today’s Web and mobile applications are designed to support large number of concurrent users by spreading load across a collection of servers. Traditional relational databases are facing many challenges to contend with these trends. NoSQL (Not Only SQL) databases are developed as a solution to these problems. NoSQL is an umbrella term for all data stores that do not follow RDBMS principles. They do not require fixed schema nor do they use the concept of joins. NoSQL databases are horizontal scalable, flexible to handle semi-structured data and support high availability. NoSQL databases are classified into four classes: Key-value data store, Columnoriented data store, Document database and Graph database. These data stores are appropriate for different applications. Key-value data stores are used to handle massive loads. Column-oriented data stores are used to store large amount of data which is distributed over many machines. Document databases are used to store semistructured data, where as graph databases are used to store data elements interconnected by many relations. This thesis provides an in-depth knowledge of NoSQL classes, and different tools available of these classes. With the social media posts and multimedia, the need of unstructured data storage has grown rapidly. MongoDB document databases are used to overcome these problems. MongoDB stores the data in the form of BSON documents. In this work an approach has been proposed for transformation of data from MySQL relational database to MongoDB document database. Further, the proposed approach has been implemented using NetBeans IDE and Pentaho.