IITiS PAN participation in the developement of national infrastructure for large-scale scientific computing

EuroHPC PLIn the years 2021-2023, the Institute of Theoretical and Applied Informatics of the Polish Academy of Sciences will participate in the implementation of the EuroHPC PL project. The project aims to provide infrastructure for advanced scientific computing, including using quantum technologies. The work carried out at IITiS PAN will focus on providing tools supporting the programming of quantum computers.

The EuroHPC PL project is used to carry out in silico research thanks to the construction and provision of specialized computing resources. This will allow large-scale calculations to be carried out directly in the framework of the EuroHPC JU joint venture - a European system of exascale-class supercomputers, also built on the basis of technologies developed in Europe. This will enable, among others simulations, analysis of big scientific data and advanced visualizations, which are the basic tools in scientific research. The EuroHPC PL project is the Polish element of the EuroHPC JU infrastructure. The project will be implemented by the Consortium composed of:

  • Akademia Górniczo-Hutnicza w Krakowie,
  • Instytut Chemii Bioorganicznej PAN,
  • Politechnika Gdańska,
  • Politechnika Wrocławska,
  • Narodowe Centrum Badań Jądrowych,
  • Instytut Informatyki Teoretycznej i Stosowanej PAN,
  • Centrum Fizyki Teoretycznej PAN.

The task of IITiS PAN is to provide a platform enabling classic simulation of the architecture of selected quantum computers. The platform will consist of several special purpose components. The most important ones include: i) an exhaustive search simulator for relatively small systems, checking all possible configurations of the problem under study (and selecting its optimal solution), ii) an emulator based on advanced numerical methods, within the range of which there are large systems characterized by relatively low entanglement and iii) a simulator that uses modern achievements of generally understood artificial intelligence (including machine learning), which is a bridge between the approaches described above.

The second task is to prepare a platform that serves as a graphical bridge between the quantum processor and the user. According to the authors' intention, such a solution will allow for intuitive and quick interaction of a potential recipient with the dynamically developing technology of hybrid classical-quantum computing.