Avaliação das Ferramentas do Big Data – Caso de Estudo: Spark Vs Flink

This article aimed to compare the performance of Big Data tools, Spark and Flink, considering five attributes that make these systems highly complex and demanding in terms of processing. As a result, the tools used to work with this data tend to be significantly more robust than conventional ones, often being more expensive as well. Open-source systems provide access to the source code, making it easier for collaborators to understand the systems and algorithms, and allowing them to adapt them according to their project needs. The methodology adopted in this study is based on descriptive research to relate the variables, with an exploratory and explanatory approach of a qualitative nature. A case study is presented, comparing the Spark and Flink platforms, considering factors such as scalability, data storage, complexity, and implementation options.