Big data is a term for data sets whose size is such large that it is difficult to process them in a reasonable time with traditional database tools or applications. Big data is typical for three things:
- Size or volume of data sets is about the size of tens of terabytes (TB) to petabytes (PB).
- Great variety of different types of data, especially unstructured data, metadata, geo-data.
- High velocity by which the data are generated
Big data is generated by the demanding scientific computations and measurements and also due to the eruption of digital content in recent years (especially the increased use of websites, applications, social networks, mobile devices and other technologies). Nowadays, we create large amounts of data with normal activities such as browsing, communication on social networks or in chat rooms, shopping at the web shop, payment by credit card, at almost any use smart phones or creating and sharing multimedia content such as photos, video or music. This trend will be in the future even more accelerated by the onset of the Internet of Things.
Big data in practice: Big data supplements traditional sources of data (originating in ERP, CRM and other enterprise systems) and extends the range of data for decision-making and management of enterprises and organizations. It helps to obtain data about the customer or customer segment or helps to find their behavior patterns. In pursuit of big data, however, we must not forget the large amount of useful data that are in “traditional” enterprise applications.
Processing and analyzes of big data are used especially in areas with growing volumes of data from operational systems. Typical examples are chain stores (data about customers, their buying habits), banking, mobile, website, online services and web applications operators, or in the field of financial and capital markets.
Furthermore, big data is in the areas of science, research or health care, where the huge amount of data from a variety of measuring instruments is penetrated and where it can be analyzed current and historical data to create a connection or use it for a simulation.
Big data is usually not centrally stored and structured. There is a large number of it and there are rapidly penetrated new ones. Business intelligence technology can be used for its analysis, but with the restriction that the results are not immediately and also scaling of BI technologies for the needs of big data would be too expensive. Big data technologies can process enormously volumes of data without the need for enormous growth of capacity and processing power of servers.