A Thesis Submitted to the School of Graduate Studies in Partial Fulfilment of the Requirement for the Degree of Master of Science in Software Engineering

Abstract

The pervasive influence of technology has transformed various features of our lives, including the area of education. Online examination has become a popular alternative to traditional paper-based tests, offering flexibility and convenience. Ethiopian public universities have also shifted to web-based online exams to improve efficiency, accessibility, security, and system performance. However, the existing system faces challenges in terms of system performance and security, which require improvement. Such issues and challenges were firstly clustered into the selected parameters in the current online examination system. After that, this research investigated and analyzed the primary data collected using surveys, interviews, and observation to discover the hidden insights to develop a new solution framework for online examination systems. This study assessed the performance and security aspects of the existing design artifacts of the online exam systems and then suggested the enhancing strategy of the performance and security-related parameters such as availability, reliability, scalability, response time, accessibility, load balancing, and others by developing a cloud-based security and performance enhancement framework for Ethiopian public universities. In addition, this research explored cloud computing as a new paradigm to resolve the current challenges and to provide alternative platform for online examination systems as a new paradigm. The researcher used exploratory and applied research design and methodology with a mixed (qualitative and quantitative) data analysis approach. The proposed and validated framework with a demonstrable functional prototype and accepted by professional users through user acceptance revealed that a cloud-driven framework for assuring security and performance for online examinations can be instrumental in optimizing availability, response time, load balance, horizontal and vertical scalability, and reliability.