We’ve been discussing posts one and two of this series, the massive influence of Big Data and why it’s critical for businesses to develop a strategy to leverage this rich trove of information. Estimates are that 90% of the world’s data was created in the past 2 years, and that rate will continue to grow exponentially. Smart businesses realize that Big Data is an opportunity to turn more data into deeper insights. The first step on that journey is to become familiar with the terrain and some of the toolsets that will help your business mine these insights. In the last couple of articles we discussed the ever growing and popular Big Data processing platform known as Hadoop. But often along with Hadoop, we hear about another toolset called NoSQL. So let’s take a look under the hood and see what exactly NoSQL is and how it helps us to access and manage Big Data insights.
Relational No More . . .
NoSQL is a technical term that means either “No SQL” or “Not only SQL” and began to emerge in recent years as a way to describe a deliberate shift among web developers away from a sole reliance on relational databases. Relational databases (RDMS) emerged in the 1970s and were based on a set of data tables that could be queried and matched based on languages like SQL. These database architectures were ‘structured’ meaning that the data was organized in a uniform format and varied little over time.
However, the massive growth of the web and social media since the early 2000s has highlighted the sheer impossibility of fitting all data into one database table on one machine. Developers therefore have looked to alternative methods for managing the huge amounts of structured and unstructured data (Facebook now stores 100 petabytes of data online in Hadoop).
By addressing the traditional one-size-fits-all approach to database management, NoSQL is an alternative framework that provides solutions for the massive scalability required with real time Big Data processing. Significantly, NoSQL systems are also referred to as “Not only SQL” to emphasize that they also allow SQL-like query languages to be used.