Search papers, labs, and topics across Lattice.
This study introduces EduMPI, a user-friendly learning support tool designed to simplify the complexities of parallel programming education by automating cluster usage and providing real-time visualizations of MPI communication. The comparative usability study shows that EduMPI significantly lowers entry barriers for students, enabling them to better understand performance analysis concepts without the steep learning curve associated with traditional tools. Key results indicate that students using EduMPI exhibited a marked improvement in their ability to identify performance issues compared to those using established professional tools.
EduMPI transforms the learning experience in parallel programming by making performance analysis accessible and intuitive for students.
Parallel programming curricula encompass not only the development of parallel code and algorithm design but also emphasize efficiency, optimization, and performance analysis. To equip students with the skills necessary for writing efficient parallel code using message passing with MPI, practical experience on HPC environments is essential. Performance analysis tools assist in identifying issues such as load imbalances or bottlenecks. Despite their use by experienced developers, these tools' complexity and required knowledge of cluster architectures, resource management, MPI, and common parallel issues hinder their educational integration. To address these barriers, we developed EduMPI, a learning support tool designed to simplify cluster usage and performance analysis for students. EduMPI offers an intuitive GUI that automates program execution on clusters and delivers near-real-time visualizations of MPI communication. This enables students to track process communication according to their physical placement within the cluster and detect performance problems interactively. This paper presents a user study comparing EduMPI with established professional performance analysis tools, demonstrating that EduMPI lowers entry barriers and fosters an intuitive understanding of parallel program performance, thereby enhancing its educational value.
