A Scalable Operating System Experiment Platform Supporting Learning Behavior Analysis

AbstractContribution: The design of an operating system (OS) experiment course with a gentle learning curve is proposed and a scalable OS experiment platform supporting learning behavior analysis is presented. Background: In the teaching practice of the OS experiment course, several problems were faced. First, the learning curve for the students is too steep. Second, the OS experiment is hard to scale to a large number of students. Finally, it is difficult to track the learning behaviors of the students and to provide feedback to the teaching plan. Intended Outcomes: By smoothing the learning curve and providing targeted and rapid feedback, students using the system can more effectively master the knowledge of the OS by completing experimental tasks. Application Design: A large number of students can log in to the experiment system and complete multiple labs online. In addition to using automated testing to provide fast feedback, the lab system also collects learning behavior data of the students, enabling the teachers to analyze the data and adjust their teaching plan accordingly. Findings: The performance and learning-behavior data of the students from 2015 to 2018 showed that the OS experiment is effective to scale to a large number of students with satisfiable teaching effectiveness.