Energy consumption has become crucial in every aspect of our lives. Researchers proposed how to cut down energy usage in various fields, specifically in computer science. New solutions for green data centers as well as placing energy-aware software within each architectural layer from the cloud through the edge to the device may result in visible energy reduction and more energy-efficient processing. In this article, we present results of performance and energy investigation of web/server request processing scenarios implemented in Spring framework using imperative and reactive paradigms respectively. Both approaches were tested using various application models – request processing with database operations as well as request processing that involves compute-intensive operations. Tests were performed for both static and dynamically changing client load. We investigated how different paradigms affect performance and energy usage and determined that while in all considered cases the reactive processing framework consumes more energy per request it provides almost faultless processing and continuous stability under a growing load.