Software Atelier: Simulation, Data Science & Supercomputing
People
Course director
Description
In this software atelier, students will select one of the proposed semester projects, designed by the course director in collaboration with research or industrial partners, and work in potentially interdisciplinary teams to address research questions. The proposed topics will be centered around:
- Supercomputing: Students will explore the architecture and operational principles of supercomputers, optimizing and parallelizing algorithms for high-performance computing (HPC) environments.
- Data science and machine learning: Students will analyze and interpret large-scale data sets, employing machine learning and statistical methods to derive meaningful insights.
Objectives
This project-based course is designed to equip students with cutting-edge skills in high-performance computing, simulation studies, and data science, and to help them tackle complex computational challenges across various scientific and engineering domains.
Teaching mode
In presence
Learning methods
- Project-based learning: Students will participate in collaborative projects addressing real-world computational problems, in particular the EUMaster4HPC student challenge in Fall 2025. Group work will be encouraged on a per-project basis.
- Preparation of scientific reports: The expected outcome of each semester project is a scientific report that will be published in the Technical Reports series of USI, or presented as a publication or poster in top-tier ACM/IEEE Swiss-based conferences [1, 2, 3]. The final grade will be based on the evaluation of these reports.
- Expert guidance: Students will be closely guided by academic researchers from USI and collaborating institutions, and depending on the project’s nature, by industry professionals.
- Eligibility: The course is available to MSc students from the EUMaster4HPC program. Students from other programs might be accepted after the approval by the course directors.
[1] ACM Platform for Advanced Scientific Computing (PASC), https://pasc25.pasc-conference.org/
[2] Data Science for the Sciences, https://www.ds4s.ch/
[3] IEEE Swiss Conference on Data Science, https://sds2025.ch/
Examination information
This is a project-based class - the grade is composed of a final project report and/or a poster that will be reviewed and discussed with an oral exam during the official exam period.
Bibliography
- ACM Platform for Advanced Scientific Computing (PASC) (The Platform for Advanced Scientific Computing (PASC) Conference is an interdisciplinary conference in HPC that brings together domain science, applied mathematics and computer science – where computer science is focused on enabling the realization of scientific computation. The PASC Conference provides three days of stimulating and thought-provoking technical sessions, including keynote presentations, minisymposia, peer-reviewed papers, panels and poster sessions. The conference is co-sponsored by ACM SIGHPC, and full papers are published in the ACM Digital Library.)
Education
- Master of Science in Computational Science, Atelier, Elective, 2nd year
Prerequisite
- Effective High-Performance Computing & Data Analytics, Holt T. A. B., Lechekhab M., Schenk O., SP 2022
- High-Performance Computing, Schenk O., Holt T. A. B., Lechekhab M., SA 2021-2022
- Introduction to Data Science, Wit E. J., Ceoldo G., SA 2021-2022
Study trips
- Swiss National Supercomputing Centre, 16.10.25 - 16.10.25 (Compulsory)
