Just as no single definition of Big Data exists, no specific cause exists for what’s behind its rapid rate of adoption. Instead, several distinct trends have contributed to Big Data’s momentum.
New data sources
Today, we have more generators of information than ever before. These data creators include devices such as mobile phones, tablet computers, sensors, medical equipment, and other platforms that gather vast quantities of information. Traditional enterprise applications are changing, too: e-commerce, finance, and increasingly powerful scientific solutions (such as pharmaceutical, meteorological, and simulation, to name a few) are all contributing to the overall growth of Big Data.
Larger information quantities
As you might surmise from its name, Big Data also means that dramatically larger data volumes are now being captured, managed, and analyzed.
New data categories
How does your enterprise’s data suddenly balloon from gigabytes to hundreds of terabytes and then on to petabytes? One way is that you start working with entirely new classes of information. While much of this new information is relational in nature, much is not. In the past, most relational databases held records of complete, finalized transactions. In the world of Big Data, sub-transactional data plays a big part, too, and here are a few examples:
Relational databases and associated analytic tools were designed to interact with structured information — the kind that fits in rows and columns. But much of the information that makes up today’s Big Data is unstructured or semi-structured, such as these examples:
Commoditized hardware and software
The final piece of the Big Data puzzle is the low-cost hardware and software environments that have recently become so popular. These innovations have transformed technology, particularly in the last five years. As we see later, capturing and exploiting Big Data would be much more difficult and costly without the contributions of these cost-effective advances.