High-Performance Computing in Practice

Scalability is the key to groundbreaking innovations – from numerical simulations to AI training on supercomputers. This course offers a hands-on introduction to the world of High-Performance Computing: code is optimized, computations are accelerated with GPUs, and HPC can be experienced up close and personal.

• Fundamentals of High-Performance Computing: architecture, parallelization, and efficiency improvements
• Code and algorithm optimization: boosting performance through better data and computation structures
• Computing with CPUs and GPUs: differences, applications, and challenges
• Working with distributed systems: data management, communication, and scaling
• Project work: independently implementing an HPC project with practical application

• Students can explain the fundamental concepts and principles of High-Performance Computing, including parallelization and cluster architectures.
• They analyze the performance requirements and scalability of algorithms in HPC environments.
• They evaluate different strategies for optimizing code and data access in distributed and GPU-based systems.
• They develop efficient solutions for compute-intensive applications, taking into account both hardware and software aspects.
• They apply their knowledge in a hands-on HPC project and document their results in a structured and comprehensible manner.

• Parallel Computing (pac), parallel enrollment possible

This module is conducted in collaboration with a research or industry partner and includes a theoretical part at FHNW as well as a practical part at the partner institution. Direct interaction with researchers and industry profes-sionals provides insights into real-world challenges and practical use cases in the field of High-Performance Computing. The knowledge gained is then deepened and applied in an independent project.