High-Performance Computing for the Future – Efficient Algorithms and Hybrid Architectures for Data-Intensive Applications
Research focus areas
We develop high-performance algorithms for data- and compute-intensive applications in science, engineering, and industry. A central focus is the efficient parallelisation of programs on distributed systems — across clusters, cloud infrastructures, and specialised HPC environments.
A particular strength lies in the close integration of High-Performance Computing (HPC) with Artificial Intelligence methods. We combine classical numerical approaches with modern AI workloads, for example in image processing, the simulation of physical processes, or the analysis of large-scale scientific datasets. By leveraging hybrid architectures (CPU/GPU), we aim to achieve maximum performance while ensuring high energy efficiency.
Our expertise includes:
- GPU-optimised algorithms and data pipelines
- AI-accelerated simulation and modelling
- Scalable training and inference environments for modern models
- Design and operation of robust, production-grade HPC systems
With our own high-performance computing system, we operate a powerful infrastructure that reliably supports research and industry projects — including compute-intensive AI experiments and large-scale simulations.
HPC Lab
High Performance Computing for data- and compute-intensive research and industry-oriented projects
Offering
We provide expertise and infrastructure in the field of High Performance Computing (HPC) for demanding applications, particularly in Artificial Intelligence, Machine Learning, Data Science, and simulation-based methods. This includes the use and exploration of novel HPC accelerators (e.g. GPUs, DPUs, and specialized accelerator architectures), as well as energy-efficient and sustainable computing with low-energy devices.
The HPC Lab supports research and development projects in the efficient use of modern computing architectures, the scaling of algorithms, and experimentation with new, resource-efficient technologies — from prototyping to production-ready scenarios.
Contact us
Prof. Dr. Tomasz Kacprzak
- Phone
- +41 56 202 79 49
- tomasz.kacprzak@fhnw.ch
Infrastruktur
- High-performance compute cluster with modern multi-core processors and GPU accelerator
- Resources for energy-efficient and low-energy computingDevelopment and testing environment for scalable algorithms and HPC workflows
- Tools and frameworks for performance analysis and optimization
- Support for integrating new accelerator technologiesHigh-bandwidth network architecture with interconnects of up to 400 Gbit
- Cluster playground for replicating different system architectures, software stacks, and orchestration environments
- Specialized hardware such as Apex Storage X21 (storage acceleration), FPGA-based compute nodes, and other domain-specific accelerators
Contact us
For further information about the FHNW School of Computer Science or to discuss potential collaboration opportunities, please contact us:
Prof. Dr. Tomasz Kacprzak
- Phone
- +41 56 202 79 49
- tomasz.kacprzak@fhnw.ch
