The recent explosion of multimedia data masses, and primarily video, generated by both professionals and individuals, forces us to review traditional analysis methods and mechanisms. Indeed, the database as we know it today is no longer just SQL (NoSql)[1]. It must be able to store and organize completely heterogeneous data on increasingly complex models, with the additional constraint of size… Volumetrics.
In this context, we see a word, an expression emerging,
“Big Data” is often defined by the “3Vs” rule[2]: “…high-volume, –velocity and -variety information assets that demand cost-effective, innovative forms of information processing for enhanced insight and decision making…”, this defines the challenges inherent in the growth of data as: Volume, Velocity and Variety. From now on, up to “5 Vs” are attributed to the characteristics of Big Data: Veracity (a concept put forward by IBM) which discusses relevance by taking context into account. Value, which instead shows the economic, scientific and technological interest of the data.
These 5 characteristics, axes or dimensions of Big Data are taken into account to succeed in creating value where traditional tools were constrained by data volume and types. Today, Big Data helps promote preventive and personalized medicine. For example, the analysis of search engine queries has already allowed for faster detection of the arrival of a flu epidemic. In the near future, connected devices should allow continuous analysis of patients’ biometric data. It also allows, through the analysis of Big Data (data from public transport passes, geolocation of people and cars, etc.), to model population movements in order to adapt infrastructure and services (train schedules and frequency, for example). As shown by the profile of participants at the Big Data Paris 2014 trade fair, almost all economic sectors are concerned by the use of this technology.
However, it is important to know that the deployment of Big Data solutions has a cost, which is financial (infrastructure, software solutions, etc.), usage-based (training and re-skilling of employees), security-related (by centralizing data, a highly coveted target is created), but also related to personal data protection (a major issue at the European level).
From experience, it should be noted that companies today have the will to do Big Data, but we quickly realize that they do not have the need and that traditional solutions meet them well. Indeed, they do not match the “5 Vs”, sometimes because of the volume which in the end is not as significant, or the velocity which is not as important.
To conclude, “Big Data” is a new disruptive technology for some, a logical evolution of Business Intelligence (BI) tools for others. It is necessary to study, analyze and understand one’s needs upstream before any migration or deployment. Then, know how to urbanize the infrastructure with the constraints encountered in the upstream phase. One should not hesitate to be accompanied in these operations by increasingly specialized ESN (Entreprise de Services du Numérique)[3].
sources
[1] https://fr.wikipedia.org/wiki/NoSQL%3Dsike
[2] «Gartner’s Big Data Definition Consists of Three Parts, Not to Be Confused with Three “V”s» 2013
[3] http://www.syntec-numerique.fr/content/les-ssii-changent-de-nom-et-se-renomment-esn
