Category: General

What can you expect?

Meet Perun – Experience a Supercomputer Up Close

During a guided tour, you will visit the containerised data centre housing the PERUN supercomputer. You will learn how it works, what kinds of tasks it can handle, and why high-performance computing is becoming an increasingly important tool not only in scientific research but also in industry and business.

From Supercomputing to Business Applications

Discover how high-performance computing (HPC) and artificial intelligence can support product development and testing, simulations, process optimisation, and large-scale data analysis. Our experts will explain how these technologies can help address specific challenges faced by your company.

Coffee & Connections

Following the tour, you are invited to an informal networking session over coffee and light refreshments. This will be an opportunity to ask questions, discuss your ideas, and meet representatives of the National Competence Centre for HPC and the Computing Centre of the Slovak Academy of Sciences to explore potential support and collaboration opportunities.

When and Where?

Date: 17. november 2026 (utorok)

Time: Choose from two available sessions:

• 10:30–11:30
• 13:30 – 14:30

Venue: Slovak Academy of Sciences, Dúbravská cesta 9, Bratislava, Slovakia

Admission: Free of charge, with prior registration required.

Each session will last approximately 60 minutes and will take place in a small group to ensure plenty of opportunities for questions and personal discussions.

Who Should Attend?

Perun Business Day is designed for entrepreneurs, managers, innovators, and representatives of small, medium-sized, and large companies interested in exploring advanced technologies and their practical business applications.

No technical knowledge or previous experience with supercomputers is required. All you need is an interest in innovation and a willingness to discover new opportunities for your business.

Registration

Participation is free of charge, but places in each session are limited. We therefore recommend registering in advance and selecting your preferred time slot.

REGISTER NOW

Open OnDemand: Access Both Supercomputers with Just a Few Clicks

Open OnDemand (OOD), our user-friendly, web-based interface for accessing HPC resources, is now available on Perun as well as Devana.

Open OnDemand makes working with HPC systems much more accessible, especially for users who are less familiar with Linux commands or traditional job submission workflows.

Through a simple browser interface, you can:

• Access both Devana and Perun without the need for an SSH client.
• Submit computational jobs with just a few clicks, using convenient web-based forms.
• Monitor and manage your running and queued jobs.
• Browse, upload, download, and manage files directly through your browser.
• Access interactive applications as they become available.

We are continuously working on expanding the range of applications and features available through Open OnDemand to make your HPC experience even smoother.

We encourage all users to give it a try! Whether you are an experienced HPC user or just getting started, we would love to hear your feedback. If you have suggestions for additional applications, features, or improvements, please let us know. Your input will help us tailor the service to your needs.

For technical information and instructions on running AMS on Devana, visit our AMS user documentation.

Amsterdam Modeling Suite (AMS) Now Available on Devana

We are also pleased to announce that the Amsterdam Modeling Suite (AMS) is now available on the Devana supercomputer.

AMS is a powerful computational chemistry and materials modeling software suite that provides a broad range of simulation methods and tools for studying molecules, materials, and their properties.

Its applications span several research areas, including:

• Computational chemistry: molecular structures, reaction mechanisms, chemical bonding, and reaction pathways.
• Materials science: electronic, structural, and mechanical properties of solids, surfaces, and nanomaterials.
• Catalysis and surface chemistry: catalytic reactions, adsorption processes, and interactions at surfaces and interfaces.
• Molecular dynamics: simulations of atomic and molecular motion, including reactive processes.
• Spectroscopy: prediction and interpretation of molecular and material properties relevant to experimental spectroscopy.
• Energy-related research: modeling of materials and processes relevant to batteries, solar cells, and other energy technologies.

AMS integrates a variety of computational methods, including Density Functional Theory (DFT), semi-empirical quantum chemistry, tight-binding methods, molecular mechanics, and molecular dynamics. Its different modules allow researchers to select methods appropriate to their systems, from detailed electronic structure calculations to simulations of larger and more complex materials.

Interested in Exploring AMS? Let Us Know!

To help researchers make the most of the software, we will arrange a dedicated workshop in cooperation with the AMS developers in the future.

This event will introduce different AMS modules, demonstrate practical workflows, and provide hands-on insights into selecting and applying suitable computational methods for specific research problems.

If you are interested in attending, or have particular topics or modules you would like to explore, please get in touch with our HPC team. Your feedback will help us shape the workshop around users’ interests and research needs.

For technical information and instructions on running AMS on Devana, visit our AMS user documentation.

Simulating Processes Inside Batteries

Scientists at Lawrence Livermore National Laboratory (LLNL) are investigating advanced modelling approaches for materials used in rechargeable batteries, with particular attention to single-crystal materials. These materials are of interest because their crystal structure influences their mechanical and electrochemical properties. Computational simulations help researchers investigate the relationships between material structure, behaviour during charging and discharging, and overall battery performance. This makes it possible to better understand the mechanisms contributing to material degradation and limiting battery lifespan.

Computational modelling does not replace laboratory experiments, but it can significantly complement them. It enables researchers to explore different material properties and operating conditions before conducting expensive physical tests, allowing experimental work to focus on more promising approaches. Advanced simulations also help connect knowledge across different scales, from processes within material structures to the behaviour of complete battery systems.

Computing Technologies Supporting Energy Innovation

These approaches highlight the importance of high-performance computing technologies for materials science and industrial innovation. The ability to simulate complex physical and chemical processes can accelerate the development of materials tailored to specific requirements. In the longer term, such research could contribute to improved battery technologies for electric mobility, portable electronics and large-scale energy storage.

Source: HPCwire / Lawrence Livermore National Laboratory, 23 February 2026

https://www.hpcwire.com/off-the-wire/llnl-advanced-simulation-and-modeling-pave-a-path-forward-for-single-crystal-battery-materials

A New 150-Qubit System

The next expansion phase of Euro-Q-Exa has begun at the Leibniz Supercomputing Centre in Germany. A new 150-qubit quantum computer will complement the existing 54-qubit system.

Both systems are based on superconducting quantum technology. IQM Quantum Computers is responsible for supplying and installing the new system.

Following installation, cooling, calibration and testing, the new equipment will be integrated into the LRZ infrastructure. Access for users is planned during 2027.

Computing Resources for Research and Industry

Euro-Q-Exa is not intended exclusively for academic research. Companies can also apply for free computing time to explore potential applications of quantum technologies.

Researchers and businesses from across Europe can request access through the EuroHPC Joint Undertaking.

Interest in the infrastructure is already growing. Since mid-September 2026, more than 37 teams from different scientific disciplines have been using Euro-Q-Exa's quantum computing resources.

Quantum Technologies Within Europe's Computing Ecosystem

Euro-Q-Exa forms part of Europe's broader effort to develop quantum computing capabilities and integrate them with high-performance computing infrastructure.

The project is funded by the European Commission, the relevant German federal ministry for research and technology, and the Bavarian Ministry of Science.

The expansion represents another step towards making advanced computing technologies accessible to European research institutions and businesses without requiring them to build expensive infrastructure of their own.

Source: HPCwire / LRZ, 12 October 2026

https://www.hpcwire.com/off-the-wire/lrz-euro-q-exa-enters-next-expansion-stage

Climate Simulations at Approximately Five-Kilometre Resolution

The Climate Change Adaptation Digital Twin (Climate DT) has produced new simulations representing global climate processes at a spatial resolution of approximately five kilometres. Compared with traditional global climate models operating at resolutions of around 100 kilometres, this represents a significant advance.

This level of detail allows researchers to better investigate events affecting specific regions, including extreme rainfall, flooding and heatwaves.

The newly released datasets include simulations of climate evolution from 1990 to 2049, as well as scenarios exploring how weather events observed between 2017 and 2025 might unfold under different climate conditions.

Powered by European Supercomputers

Producing high-resolution climate simulations requires enormous computing resources. The project therefore relies on infrastructure supported by the EuroHPC Joint Undertaking, particularly LUMI in Finland and MareNostrum 5 in Spain.

The Climate DT is developed by a European partnership led by the Finnish supercomputing centre CSC in close collaboration with the European Centre for Medium-Range Weather Forecasts (ECMWF).

Supporting Better Decisions

The simulations can help inform decisions in areas such as water management, renewable energy and wildfire risk assessment.

The climate datasets and accompanying user guidance have been made available through the Destination Earth platform to researchers and experts from both the public and private sectors.

The project demonstrates that high-performance computing is not limited to fundamental research. By enabling detailed simulations, HPC can provide valuable information for practical planning and help societies prepare for changing climate conditions.

Source: HPCwire / CSC, 27 May 2026

https://www.hpcwire.com/off-the-wire/new-destination-earth-climate-digital-twin-simulations-released

The publication introduces a methodological framework for identifying and evaluating research results with potential for further exploitation, known as Key Exploitable Results (KERs). It emphasises that not every technically successful outcome is automatically ready for practical application. Factors such as its potential benefits, target users, existing alternatives, and intellectual property rights must also be considered.

The authors propose a five-stage evaluation process, from identifying potentially exploitable results to assessing their readiness for practical implementation. The methodology applies to software tools, computational methods, datasets, services, and training materials.

The report was developed through collaboration between experts from six European HPC Centres of Excellence: BioExcel, HiDALGO2, EXCELLERAT P2, MultiXscale, ESiWACE, and ChEESE-2P.

It is the first in a series of four publications that will explore research exploitation strategies, readiness assessment, further development, and long-term sustainability.

The publication is intended for researchers, developers, project managers, and innovation professionals involved in translating scientific research results into practical applications.

More information:

Article on the HPC in Europe portal

Scientific publication – DOI

12 October 2026, 14:00–15:00 | Getting Started with EFP

This webinar will cover the first steps in using the platform – from creating an account through MyAccessID, identity verification and setting up multi-factor authentication (MFA), to accessing allocated EuroHPC resources. It will also explain how a Principal Investigator (PI) can manage a project and how project members can gain access to allocated resources.

Register for Getting Started with EFP

19 October 2026, 14:00–15:00 | EFP Features Showcase

A practical overview of the platform’s key features, including the management of EuroHPC projects and allocations, access via SSH certificates, the federated software catalogue based on EESSI, the interactive web interface, and workflow management.

Register for EFP Features Showcase

26 October 2026, 14:00–15:00 | EFP Workflows

The final webinar in the October series will focus on workflows within the EuroHPC Federation Platform. Further programme details will be announced soon.

Register for EFP Workflows

Easier Access to European HPC Resources

The EuroHPC Federation Platform is designed to gradually simplify access to and use of Europe’s supercomputing infrastructure by connecting services that have previously been available through different systems and interfaces. The webinar series is therefore relevant not only to current users of EuroHPC resources, but also to researchers and other potential users who are just beginning to explore the opportunities offered by Europe’s HPC infrastructure.

When a Trading Algorithm Learns to Adapt

Traditional automated trading systems operate according to predefined rules. However, financial markets are constantly changing, and an algorithm that performs well during one period may gradually lose its effectiveness.

As part of the EuroCC project, Smart-Markets therefore explored whether machine learning – specifically reinforcement learning – could be used to create a system capable of adapting its decisions to changing market conditions.

More than ten years of high-frequency EUR/USD market data, capturing individual price changes, were used to train and subsequently test the AI agent. Processing such large volumes of data and running repeated experiments is also a good example of a task where high-performance computing infrastructure can provide significant benefits.

From Experiment to Proof of Concept

The results showed that the AI agent gradually learned to make its own trading decisions and, in a simplified test scenario, was able to adapt to changing market conditions even during periods of negative market development.

At the same time, the project also revealed the limitations of the current solution. The test scenario did not include transaction costs, and under more realistic conditions, it has not yet been possible to train a system capable of generating a profit. The result is therefore a Proof of Concept rather than a ready-to-use investment tool. Further development could include, for example, broader data inputs and more complex modelling of real-world trading conditions.

HPC as a Tool for Small and Medium-Sized Enterprises

The Smart-Markets example also demonstrates one of the key roles of the EuroCC network – making high-performance computing accessible to companies with no previous experience in HPC.

Through the Proof of Concept, Smart-Markets was able to experiment with a new AI technology without having to build its own computing infrastructure or possess in-house AI expertise from the outset. HPC resources made it possible to process and analyse large volumes of high-frequency financial data and test new approaches to adaptive trading systems.

Connecting companies with computing infrastructure, expert knowledge and support for developing new solutions is one of the key roles of the National Competence Centres within the EuroCC network.

Listen to the Full Interview

The episode AI trading bots – Marc Pötter is approximately 45 minutes long. The interview is hosted by Andreas Lintermann from the Jülich Supercomputing Centre and, in addition to the experiment itself, explores the company’s practical experience with HPC and the opportunities that high-performance computing can offer small and medium-sized enterprises.

Listen to the episode AI trading bots – Marc Pötter

Read the Smart-Markets success story on the HPC in Europe Portal