Workshop: Increasing Undergraduate Research in Computational Sciences
Abstract: High Performance Computing (HPC) is used everywhere from e-commerce to machine learning. The main problems solved by HPC are related to computational sciences and affect almost all scientific disciplines. Training the students that will have sufficient knowledge in HPC is a crucial task of computer science departments in all educational institutions. This paper describes the steps taken to increase the undergraduate engagement in HPC disciplines in our undergraduate institution. The increased exposure allowed our computer science students to collaborate with other departments research projects by accelerating different computational intensive problems. In order to increase students knowledge in HPC we decided to offer classes at the undergraduate level in parallel programming and distributed systems; and to properly support these classes we invested in a lab specifically designed for parallel programming, with machines equipped with the best multicores cpus and gpus so students can experience the effects of massively parallel architectures in their code.
Using detailed enrollment, master thesis and publications from the last few years, since the last has been available; we can see a substantial increase in undergraduate students interest in the HPC area, and a great success in the number of research master thesis as well as the number of peer reviewed undergraduate and graduate student-authored publications.