Ostrava
The CLARA Testbed is supported by state-of-the-art core computing facilities in Ostrava, providing high-performance computing and quantum computing resources for neuroscience research.
These facilities include:
Karolina is the most powerful supercomputer for academia in the Czech Republic. It is hosted at IT4Innovations National Supercomputing Center at VSB—Technical University of Ostrava. Karolina provides exceptional computational power for large-scale simulations, AI-driven medical research, and neurodegenerative disease modelling.
The LUMI supercomputer, with a performance of 531,5 PFlop/s, is one of Europe's most powerful supercomputing systems. It is installed in Kajaani, Finland, and was a joint investment of the EuroHPC Joint Undertaking and the LUMI consortium.
The Czech Republic is also part of the LUMI consortium, thanks to the involvement of IT4Innovations.
A state-of-the-art quantum computer VLQ operated at IT4Innovations will be integrated with the EuroHPC supercomputer Karolina, enabling hybrid quantum-classical computing for scientific, industrial, and societal applications. Co-funded by the EuroHPC Joint Undertaking and a consortium of nine European countries, VLQ aims to provide cutting-edge quantum computing resources to researchers and industries across Europe.
Timeline for installations: until September 2025
It is a computational cluster focusing on optimising AI computations, especially training new deep neural networks (models) and possibly their subsequent use for given tasks (so-called inference). One of the project's main goals will be to use AI technologies to simulate quantum systems. The module brings together three central units providing both the computation itself, the storage and processing of operational data, and last but not least, a dedicated network connecting computational and data capabilities: a GPU computing cluster, a robust SCRATCH data storage, and a high-speed network.
Timeline for installations: June 2026
HPCQC short-term storage capacities will be complemented by so-called “cold storage”, which provides necessary medium and long term storage capacities to the users. This storage will serve to store data in the active phase of their processing and in the case of not too high demands on throughput it can even partially replace the powerful SCRATCH storage, especially as a source of data for learning new neural network models.
This data storage layer will also serve as an intermediate link between computational capacities and data repositories, enabling the preparation and publication of the center's outputs concerning the principles of open science and FAIR data access.
Timeline for installations: June 2026
