文献类型：期刊文献

英文题名：Enhancing Education Accessibility: Portable Microservers for Computer-Based Testing in Resource-Constrained Environments

作者：Dhuny, Riyad[1]; Ying, Fangli[2]

机构：[1] University of Technology, School of Innovative Technologies and Engineering, 11134, Mauritius; [2] East China University of Science and Technology, Department of Computer Science, Shanghai, China

年份：2025

起止页码：36

外文期刊名：22nd International Learning and Technology Conference: Human-Machine Dynamics Fueling a Sustainable Future, L and T 2025

收录：EI(收录号：20251618270059)

语种：英文

外文关键词：Computer debugging - Negative bias temperature instability - Problem oriented languages - Teaching

摘要：This study explores the potential of LAMP stack open-source applications for embedded computing to enhance education accessibility, particularly in resource-constrained environments. The research focuses on converting a Raspberry Pi (RPI) device into a portable microserver hosting a Computer-Based Testing (CBT) system, specifically TCExam. The methodology involves porting TCExam to RPI hardware, testing its operability, and evaluating its performance under various user loads. Experiments were conducted using RPI versions 4 and 5, with Apache JMeter employed for performance testing. Results demonstrate that RPI4 can sustain at least 30 concurrent users, while RPI5 supports over 50 users in a load-testing environment. The study applies Menascé's equation to estimate the theoretical number of supported users, revealing the potential for microservers to accommodate larger class sizes than initially expected. The research highlights the feasibility of using affordable, portable hardware for deploying CBT systems, which is particularly beneficial in areas with limited internet connectivity or power sources. This approach offers educators a secure, offline assessment environment and the flexibility to prepare and test the system before deployment. The study concludes that microservers like RPI can increasingly run applications previously restricted to server rooms, opening new possibilities for improving education quality in developing countries. Future work suggests integrating local Large Language Models for offline use in remote areas, further enhancing the educational toolkit available to educators in resource-constrained environments. ? 2025 IEEE.