Centre for High-Performance Computing
The Centre performs quantum chemical modelling and simulations of materials and processes using a new-generation server cluster. Computational and machine learning methods are applied to predict properties and to develop new materials with applications in clean technologies and mechatronics.
Description of Activities
Description of Services
Quantum chemical modelling and prediction of the properties of a wide range of materials
Detailed description of the service: The research is carried out on a latest-generation computing server cluster, which enables:
- Atomistic molecular dynamics simulations for tracking processes over time;
- Modelling using quantum dynamics methods,
- Theoretical modelling and prediction of properties and materials using neural networks.
Development of software for modelling processes and properties
Detailed description of the service: The research is carried out on a latest-generation computing server cluster, with properties and materials being studied using machine learning algorithms.
| Services offered | Price in EUR* |
| Quantum chemical modelling and prediction of the properties of a wide range of materials | 63 |
| Atomistic molecular dynamics simulations for tracking processes over time | 63 |
| Theoretical modelling and prediction of properties and materials using neural networks | 63 |
| Study of properties and materials using machine learning algorithms where a large database is available. | 63 |
| Data interpretation | 24 |
* The prices indicated are approximate and depend on the scope of the research. They include VAT at the rate of 20%.
Equipment
Computing server cluster, which consists of:
- 18 compute servers (nodes) and one management server (management node), Supermicro;
- 40 Intel Xeon Gold 6226 processors (CPU) with a total of 480 processor cores (CPU cores);
- 12 NVIDIA V100 Tensor Core graphics processors (GPU) with a total performance of 168 TFLOPS;
- 7680 GB RAM;
Quantum chemical, molecular mechanics and molecular dynamics calculations are performed on the cluster system for model systems of various sizes – from single molecules to multimolecular systems consisting of hundreds of thousands of atoms. A wide range of methods is applied, ranging from standard density functional theory (for isolated and periodic systems) and atomistic molecular dynamics simulations to innovative approaches such as ground- and excited-state ab initio molecular dynamics, multideterminant solutions, QM/MM and others.
Fields of application: organic catalysis, quantum and computational chemistry, computer chemistry and spectroscopy, characterisation of the properties of foams, emulsions and porous materials.
Contacts
Head: Prof. Galia Madjarova, fhgm@chem.uni-sofia.bg
Access to the High-Performance Computing Laboratory
The High-Performance Computing Laboratory is part of the research infrastructure of the National Centre of Excellence “Mechatronics and Clean Technologies” and is located on the Lozenets campus. Requests for access and for joint research are therefore handled centrally — see Contacts. In addition, a full overview of the Centre’s other laboratories is available on the Infrastructure and Laboratories page.
The high-performance computing equipment was established under project BG16RFPR002-1.014-0006, funded by the Programme “Research, Innovation and Digitalisation for Smart Transformation” 2021–2027, co-funded by the European Union through the European Regional Development Fund.