By Mike Marr
Expert Author
Article Date: 2010-10-26
NoSQL has managed to slip under many noses the past decade. Other than its initial plunge into the scene, NoSQL has made little to no waves. However, these systems have been improving to the point that they are now viable alternatives to traditional database management systems. Before we get into more advanced discussion on NoSQL, let's first get educated with some basic history and terminology.
A key to understanding NoSQL's rise is to understand Brewer's (CAP) Theorem. Developed by Berkley professor Eric Brewer, this theorem divides all applications into being able to have only two of the three following characteristics:Consistency - ability to keep data integrity;
Accessibility - ability to access data at any given time by any user;
Partition Tolerance - ability to operate effectively across multiple partitions/network systems.
Database management systems (DBMS), or the software used to manage and use our databases, are dominated by the relational database model. Relational database management systems (RDMS) dominate the market because they operate with both consistency and accessibility. Until recently, this has been fine, as poor partition tolerance was not as big issue. However, with the ever increasing acceptance of cloud computing, and continued globalization via outsourcing and other means, partition tolerance has become extremely vital. The increased partitioning of information systems has put this additional importance on partition tolerance, something that RDMS does not do that well. Therefore, some NoSQL DBMS are emerging with a focus on partition tolerance, and thus becoming a very appealing option to some companies.
The importance on partition tolerance is the window that is giving NoSQL an opening to enter the market, but it also has other strengths. In fact, NoSQL does not mean that it uses some other language to interrupt instructions (although it could choose not to support SQL). Rather, NoSQL stands for "not only SQL", and the idiom focuses on breaking traditional boundaries of database management systems. Most NoSQL systems remove the multi-platform support and extra unnecessary features, making them much more lightweight than their RDMS counterparts.
Be sure to look for the upcoming NoSQL articles. We will further compare NoSQL systems with traditional database management systems, including specific examples and case studies.
