Big Data, Big Innovation: it this the new frontier for computing?
The ever expanding realm of social media, sophisticated data capture and the ‘Internet of Things’, has prompted exponential growth in the amount of online data. In fact it’s estimated that at present, we’re creating around 2.5 quintillion bytes of data every day – more than has been seen by everyone since the beginning of time. Every single day.
‘Big Data’ in computing essentially refers to the use of this data to find insights, answer questions and seek opportunities for business that might otherwise have slipped under the radar. And with so much data to now draw from, this opens a world of opportunity.
What is Big Data?
Big Data spans three main dimensions: volume, velocity and variety.
Enterprises are faced with an ever-growing expanse of data of all types. This increase in the depth and breadth of resource offers businesses a wealth of opportunity to convert information into consumer insight and ultimately, more sales.
Velocity refers to the rate of change in data and how quickly it must be used to create value. To illustrate, if businesses could access almost real-time analytics on every kind of data emanating from consumer mobile devices, they would be able to build a much more in-depth understanding of the consumer and their buying patterns.
Big Data is essentially any form of data; from text, to click streams, video, audio and more, and this can be either structured or unstructured. In fact, about 80% of online data is ‘unstructured’, much of which is social media.
Making Sense of it All
Naturally, because of its sheer scale and variability, fathoming Big Data requires a modern platform that can manage data of any type. Initially, Hadoop was created to provide an open-source framework. But by its very nature, Big Data is too vast and complex for a one-size-fits-all solution. In fact, there are two other technologies that have been designed to enable Big Data: NoSQL (not only SQL) and Massively Parallel Processing (MPP).
What about SQL?
There was a time when the prospect of Big Data had SQL Server consultancies and developers running scared. In fact, Hadoop was developed because at the time, there were no SQL engines that could manage such vast databases.
However, SQL access has the advantage of familiarity and compatibility with many existing tools. For this reason, it’s widely believed that there may be a future for the use of SQL instead of NoSQL, Hadoop or MPP. In fact, Facebook began to bridge this gap by introducing SQL database-type functionality to Hadoop when it created Hive in 2009.
It’s clear that though we’ve adopted ways and means of processing Big Data, the solution is not yet here. With the wealth of information steadily becoming available at our fingertips, Big Data unlocks huge potential for the future.
This guest blog is written by Ali Raza on behalf of http://www.interoute-iam.com. Interoute Group, is one of the UK’s premier providers of Application Managed Services, Managed Cloud Services and Infrastructure Solutions for Enterprises and Independent Software Vendors.