.jpg)
By Márcio Arbex *
Organizations are made up of processes and changes in their environment due to market demand. Having control of both is a difficult task, due to the vast collection of data and the need for effective methods to organize it. Thus, the use of Information Technology becomes essential for the survival of companies as it provides means for the collection, storage and treatment of data from all of its processes. To support well-founded and well-planned decision-making, there are software and tools that, combined with a well-defined methodology, offer all the support to the decision-making process. Methodologies such as organization, analysis, science and use of data provide vital support for the survival of organizations.
The growing and unfair competitiveness, combined with the most varied technologies, makes information one of the most valuable assets of any organization. Having access to data in a precise, fast and efficient way is one of the great differentials that can lead to the success of companies - not only storing a large amount of data, but also evolving in the so-called “democratization of data”. The first step so that the information needed for decision making stops being subjective and becomes a solid and well-founded process is to grant access to the data in a simple, agile and unique way.
Considering the enormous amount of data available through computational resources, retrieving them in a conventional manner, such as through SQL (Structured Query Language) queries made directly in the databases, may not extract all the information that this large mass of data provides. In addition, when there are heterogeneous databases, where the ETL process (extraction, transformation, and loading) is considered a common resource.
Among the new data access methodologies, I highlight the “Data Virtualization” - a process dedicated to extracting one or more data sources in an elaborate and thorough way, consolidated in a homogeneous and virtual data layer capable of providing access through the most diverse means. It is important to highlight that Data Virtualization does not follow the traditional processes of the ETL concept, as the data is not loaded or persisted in another source. Based on this principle, costly processes such as data replication, synchronization and storage are reduced, as there are, in fact, only virtual notes for the data in their respective sources (where the data must actually exist and remain).
The data virtualization process breaks traditional paradigms since their access and availability are considered heterogeneous and complex. Considering the governance and security processes based on a previously established information architecture, data virtualization becomes essential for companies looking to improve their data and information management.
* Márcio Arbex is Director of Pre-Sales for TIBCO in Latin America
Warning: The opinion presented in this article is the responsibility of its author and not of ABES - Brazilian Association of Software Companies
EN 
PT 
