HPCQCMark: a New Modular HPC-QC Benchmarking Framework
Supercomputing centers all over the world have started integrating Quantum Processing Units (QPU) with their supercomputers. These new hybrid architectures require software stacks with specific environments, interfaces, tools,... This has led to a need for ways to quickly evaluate the readiness and the performance of these new software stacks. While benchmarking methods exist for the individual HPC or QC systems, combined HPC-QC benchmarking is still mostly a blind spot. We thus propose a new modular, easy to customize, benchmarking framework called “HPCQCMark” to evaluate HPC-QC systems and especially their software stack. We added modular components to the framework which cover relevant HPC-QC workflows and widely used QC programming environments. Finally, we perform experiments on Riken R-CCS's and TGCC's HPC-QC environments to validate the ability of the framework to evaluate the impact of HPC-QC software stacks on code execution time.
Authors
- Stéphane Vialle (ORCID: https://orcid.org/0000-0001-6336-2269)
- Mitsuhisa Sato (ORCID: https://orcid.org/0000-0003-0543-7116)
- Miwako Tsuji (ORCID: https://orcid.org/0000-0003-4709-1969)
- Brice Chichereau
- Patrick Carribault (ORCID: https://orcid.org/0009-0003-7210-0449)
Institutions
- Centre National de la Recherche Scientifique (FR)
- Commissariat à l'Énergie Atomique et aux Énergies Alternatives (FR)
- Université Paris-Saclay (FR)
- CentraleSupélec (FR)
- CEA Paris-Saclay (FR)
- RIKEN Center for Computational Science (JP)
Publication Details
- Published
- 2025-08-30
- DOI
- https://doi.org/10.1109/qce65121.2025.10285
- Primary Topic
- Distributed and Parallel Computing Systems
- Type
- article
- Field-Weighted Citation Impact
- 0.00