Performance Evaluation of Spring Boot Framework targeting pertinence for high demanding systems

Abstract

The proliferation of internet connected devices, such as smartphones, tablets, and laptops, has driven demand for distributed applications, necessitating scalable and cost-effective solutions. Cloud computing has emerged as a critical enabler for organizations to deliver services remotely, prompting the development of specialized frameworks to support distributed architectures. However, such frameworks may introduce performance overhead, particularly for high-demand services. This study evaluates the Spring Boot framework, a Java-based API development tool, by implementing a foreign exchange (Forex) server capable of processing currency buy/sell transactions online. The system leverages key framework features, including user authentication and synchronous/asynchronous messaging. To assess performance, the server is profiled using JProfiler, with latency measured by instrumenting framework-related components with time triggers. Stress testing via JMeter includes two experimental scenarios: simulating scalability under increasing user load (1,000 to 10,000 users with 500,000 total requests), and varying request intensity under fixed user load (1,000 users with 100 to 900 requests per user). Findings quantify Spring Boot’s performance overhead and its impact on throughput and latency. We discuss implications for adopting the framework in high-demand distributed systems, offering actionable conclusions for developers and architects.