High-Performance Computing Research Group

The mission of our group is to perform computing systems research for energy-efficient high-performance computing. Our specific focus is custom computing with FPGAs (field-programmable gate arrays). This technology allows the developers to create highly efficient processing architectures by tailoring the processor architecture to the needs of the application. In collaboration with the Paderborn Center for Parallel Computing (PC²) we are validating our foundational research with practical HPC applications, frequently in cooperations with research groups from computational sciences. In teaching we offer courses, seminars, projects in the areas of custom and high-performance computing. We participate in several national and transnational research projects.

Please refer to our teaching and research pages for more detailed information. 

Research

The research of our group revolves around computer architectures, design methods and applications for energy-efficient computing systems. Our main expertise is in custom and heterogeneous computing in the context of high-performance computing. In custom computing, we use FPGA-based reconfigurable hardware architectures that can be specialized to the target application, which can lead to significantly improved performance and energy-efficiency when compared with general-purpose CPUs. Our work in HPC revolves around the topic of the development of massively-scalable algorithms for scientific computing, in particular, for the domains of physics, chemistry and engineering.

Our research is primarily foundational, with an emphasis on the development of new methods, which we evaluate by simulation or prototypes targeting case studies. In tigh collaboration with domain experts at the Paderborn Center for Parallel Computing (PC2) and users of PC2 , we validate our research also with actual scientific applications. Our research has been supported by numerous third party funding agencies, such as, German Research Foundation (DFG), German Ministry for Education and Research (BMBF), and the European Commission. We also maintain collaborations with industrial partners, such as, AMD/Xilinx and Intel/Altera.

Teaching

The High-Performance Computing Group is offering lectures, seminars, project groups and Bachelor-/Master's theses. For more information on our current and past offerings, please follow the navigations links below.

Publikationen

Multi-FPGA Designs and Scaling of HPC Challenge Benchmarks via MPI and Circuit-Switched Inter-FPGA Networks
M. Meyer, T. Kenter, C. Plessl, ACM Transactions on Reconfigurable Technology and Systems (2023).
Mutation Tree Reconstruction of Tumor Cells on FPGAs Using a Bit-Level Matrix Representation
J.-O. Opdenhövel, C. Plessl, T. Kenter, in: Proceedings of the 13th International Symposium on Highly Efficient Accelerators and Reconfigurable Technologies, ACM, 2023.
Scalable Multi-FPGA Design of a Discontinuous Galerkin Shallow-Water Model on Unstructured Meshes
J. Faj, T. Kenter, S. Faghih-Naini, C. Plessl, V. Aizinger, in: Proceedings of the Platform for Advanced Scientific Computing Conference, ACM, 2023.
FPGA Acceleration for HPC Supercapacitor Simulations
C. Prouveur, M. Haefele, T. Kenter, N. Voss, in: Proceedings of the Platform for Advanced Scientific Computing Conference, ACM, 2023.
Computing and Compressing Electron Repulsion Integrals on FPGAs
X. Wu, T. Kenter, R. Schade, T. Kühne, C. Plessl, in: 2023 IEEE 31st Annual International Symposium on Field-Programmable Custom Computing Machines (FCCM), 2023, pp. 162–173.
Breaking the exascale barrier for the electronic structure problem in ab-initio molecular dynamics
R. Schade, T. Kenter, H. Elgabarty, M. Lass, T. Kühne, C. Plessl, The International Journal of High Performance Computing Applications (2023).
Shallow Water DG Simulations on FPGAs: Design and Comparison of a Novel Code Generation Pipeline
C. Alt, T. Kenter, S. Faghih-Naini, J. Faj, J.-O. Opdenhövel, C. Plessl, V. Aizinger, J. Hönig, H. Köstler, in: Lecture Notes in Computer Science, Springer Nature Switzerland, Cham, 2023.
A High-Fidelity Flow Solver for Unstructured Meshes on Field-Programmable Gate Arrays: Design, Evaluation, and Future Challenges
M. Karp, A. Podobas, T. Kenter, N. Jansson, C. Plessl, P. Schlatter, S. Markidis, in: International Conference on High Performance Computing in Asia-Pacific Region, ACM, 2022.
Alle Publikationen anzeigen

Contact

Prof. Dr. Christian Plessl

Hochleistungsrechnen

Raum X1.101
Universität Paderborn
Mersinweg 5
33100 Paderborn

Team Assistant

Michaela Kemper

Hochleistungsrechnen

Raum X1.105
Universität Paderborn
Mersinweg 5
33100 Paderborn