Análise de desempenho, otimização e refatoração do servidor radar livre

Abstract

This work aims to analyze, optimize, and refactor the server of the Radar Livre system, an open-source platform for air traffic monitoring based on ADS-B technology. The study was motivated by the need to improve the system’s performance, which faced limitations when handling numerous data collectors and simultaneous web clients. To achieve this, two performance analyses were conducted using emulation techniques and metrics such as response time, Central Processing Unit (CPU) usage, Random Access Memory (RAM) usage, and entrada/saída rate. The first analysis identified critical bottlenecks, including inadequate database connection management and excessive server resource usage. Based on these findings, optimizations were implemented, adding a geographic extent (PostGIS) in PostgreSQL, configuring Gunicorn and Nginx, and refactoring the code to enhance maintainability and efficiency. The second performance analysis demonstrated significant improvements, with increased processing capacity and reduced response times. It is concluded that the implemented changes were effective, but the adoption of asynchronous technologies and horizontal scalability are suggested for future enhancements.