Estimated reading time: 2 minutes.
The label "Big Data" is given to a project when it meets these three criteria: volume, variety and/or velocity.
CRITERIA FOR DEFINING BIG DATA
- Volume: This is the most direct association with Big Data, as it involves information systems with either massive amounts of stored information, from Terabytes (TB) to Petabytes (PB), or high rates of information ingestion in short periods of time, e.g. platform for registration of operator requests, complaints and claims.
- Variety: The diversity of sources, formats and types of information moves a data project into a domain that exceeds the capabilities of conventional business intelligence or analytics solutions.
The unification of sources, support for multiple formats and treatment of different types of information on the same platform is the promise of Big Data's value. - Speed: Large volumes and varieties of information require faster processing architectures with the promise of not only finding business value in data, but also doing it faster.
Traditional business intelligence reporting solutions require consolidation times of weeks to months depending on the size of the databases and composition of the queries. With Big Data it is possible to achieve generation times in minutes or hours.
APACHE HADOOP AS STANDARD FOR BIG DATA SOLUTIONS
In practice, the Apache Hadoop ecosystem has become the industry standard for implementing Big Data-based solutions, taking advantage of all the benefits of an open source project.
Big Data enthusiasts exploit the robustness, maturity, collaborative community and disruption of cloud computing for the integration, manipulation and deployment of models on abundant and diverse customer information.
The latest Big Data solutions support the growth of resources on demand thanks to their operation in the cloud. The deployment of the infrastructure is done according to the complexity and response times required by the business questions, without falling into the under- or over-utilisation of resources, typical of on-premise infrastructures.
It also frees the customer from the expense and inconvenience of acquiring, installing and supporting their own infrastructure, focusing attention on the real business value: the information.
