Every CxO knows there is a universal and never ending quest for faster access to more and more data. Expectations for fast data gathering pervade our everyday life. In television and movies it is common to see spies, detectives and other forensic-types instantly obtain vast amounts of cross-referenced information, complete with pictures and maps, with just a few keystrokes or mouse-clicks. However, anyone involved with using or supplying data for business decision-making knows that reality falls far short of this fictional scenario. In fact, because it is so difficult to achieve, increasing the speed and scope of data access is a continued source of revenue for many suppliers of services and technology.
One increasingly popular approach for quickly accessing the mounds of data required for decision making is to utilize what is commonly called a "columnar database." According to a recent IDC survey, almost 40% of the respondents were using, or at least aware of, a columnar database. But 62% did not know what a columnar database is or didn't know if they have one or not.
This paper is intended to introduce the concept of columnar databases, provide a working definition of them for business users, and describe where they are most applicable. The last point is crucial; it is important to explore the value proposition of columnar databases because there are situations where they are not applicable. Finally, we will present a list of considerations to use when evaluating a columnar database. Columnar databases offer a unique solution to several data problems. This document presents the basic concepts of columnar databases in a straightforward manner so anyone who is involved with data, business or technical, can understand their purpose and application.
