<![CDATA[The performance of a web page plays a crucial role in ensuring a smooth user experience and directly influences the success of digital services. This study aims to understand how a web service responds when subjected to varying traffic loads and to identify the technical factors that most significantly shape service quality. The main focus is to evaluate response time, system stability, and server resource utilization as demand gradually increases, both during peak traffic and lighter load conditions.The research applies an experimental approach using load-testing techniques with a cascading scenario, where a predefined number of HTTP requests is sent and measured at each stage. The observed variables include CPU and memory consumption, latency values (minimum, average, maximum, and standard deviation), response time, round trip time (RTT), and throughput measured through average and peak HTTP requests per second (RPS). Additionally, two user-experience indicators—Largest Contentful Paint (LCP) and Page Load Time—are analyzed to understand their relationship with perceived speed and potential changes in bounce rate. The findings show notable improvements after optimization was applied. The server’s TTFB decreased significantly from 300–399 ms to 239–247 ms. Network RTT also improved, dropping from 64 ms to between 6 and 40 ms. Overall latency declined by about 6–7% from the initial range of 757–759 ms. Even under a test scenario involving 20 virtual users and 1,200 total requests, CPU and memory usage remained stable, while peak load decreased to about 49–50% based on p95 and p99 metrics. These results indicate that implementing Memcache and MySQLTuner contributes substantially to improving application responsiveness and enhancing users’ perception of system performance.]]>
Copyrights © 2025
