Enhancing HPC Curriculum through Competitions

Volume 17, Issue 1 (March 2026), pp. 57–58

Download PDF

@article{jocse-17-1-8,
  author={Cristina Carbunaru and Sriram Sami},
  title={Enhancing HPC Curriculum through Competitions},
  journal={The Journal of Computational Science Education},
  year=2026,
  month=mar,
  volume=17,
  issue=1,
  pages={57--58},
  doi={https://doi.org/10.22369/issn.2153-4136/17/1/8}
}

High Performance Computing (HPC) supports breakthroughs in artificial intelligence (AI), data-intensive science, and engineering. At the National University of Singapore (NUS), core parallelism concepts are currently taught through courses in Parallel Computing and Concurrent Programming, with additional domain-specific exposure in courses. While these offerings build strong theoretical foundations, they leave a gap in systems-level competencies essential for deploying, optimizing, and scaling applications on real HPC infrastructure. We addressed this gap by initiating several projects meant to increase the knowledge in system-level skills for HPC. A main initiative is the participation in HPC student cluster competitions through which we integrated training in resource management, profiling, monitoring, containerized workflows, and distributed AI workloads for our selected students. This focus enables participants to bridge programming theory with operational expertise, preparing them to work effectively with clusters and heterogeneous architectures. Building on the interest around HPC competitions, the main curriculum in computer science is developing to include full-fledged HPC courses. We faced several challenges in this process, including a steep learning curve with complex systems, limited access to costly and shared cluster resources, and a shortage of instructors with up-to-date expertise. Pedagogically, bridging theory and large-scale practice is difficult, especially in the HPC context where the access to resources is remote. Therefore, sustainable curriculum development calls for a gradual expansion of teaching topics and resources, coupled with the integration of hands-on, competition-driven learning to maintain engagement. Formal HPC training enhances students' readiness for careers in computational science, promotes cross-disciplinary collaboration, and equips graduates with the advanced skills essential for solving complex challenges in AI and data-intensive fields.