Aplicabilidade e desempenho de banco de dados NoSQL em aplicações modernas: um estudo de caso em MongoDB, Cassandra, Neo4j e Redis

Abstract

The exponential growth in data volume and complexity in the Big Data era has driven the adoption of NoSQL databases, which offer high scalability, flexibility, and performance. However, the diversity of models—such as Document, Columnar, Graph, and Key-Value— presents a challenge in choosing the appropriate technology for each type of application. This work presents a comparative case study on the applicability and performance of four prominent NoSQL databases: MongoDB, Cassandra, Neo4j, and Redis. For this analysis, an application was developed in Python, using the FastAPI framework for the API and Streamlit for the visualization interface, running in a dockerized environment. Performance tests were conducted across four distinct scenarios: batch writing, complex filtered reads, data aggregation, and relationship traversal. The results demonstrate that there is no one-size-fits-all solution, and the performance of each database is intrinsically linked to its architecture and the specific workload, thus providing practical insights for the development of modern applications.