Comparative Study of NoSQL and Relational Database

Relational Database and NoSQL are competing types of database models. The former has been in existence since 1979 and the latter since the year 2000. The demands of modern applications especially in web 2.0, 3.0 and big data have made NoSQL a popular database of choice. Choosing an appropriate database model to use is an important decision that developers must make based on the features of a given database model. This paper compares the features of Relational Databases and NoSQL to establish which database is better at supporting demands of modern applications. The paper also brings out the challenges of NoSQL. Finally, the paper concludes by determining whether Relational Databases would completely be replaced by NoSQL database models. The findings revealed that, Relational Databases are based on ACID model which emphasizes better consistency, security and offers a standard query language. However, Relational Databases have poor scalability, weak performance, cost more, face availability challenges when supporting large number of users and handle limited volume of data. NoSQL, on the other hand is based on the BASE model, which emphasizes greater scalability and provides a flexible schema, offers better performance, mostly open source, cheap but, lacks a standard query language and does not provide adequate security mechanisms. Both databases will continue to exist alongside each other with none being better than the other. The choice of the database to use will depend on the nature of the application being developed. Each database type has its own challenges and strengths, with relational database lacking of support for unstructured data while NoSQL lacks standardization and has poor security. Modern applications in web 2.0, 3.0 and big data are well suited to use NoSQL but, there are still many applications that rely on Relational Databases.