International Projects

Soft and hard magnets are essential components of several devices of utmost importance in various high-tech manufacturing industries. As an example, they play a central role in the functioning of electric motors, which are a pivotal element in the ongoing transition from fossil fuels to renewable energy encouraged by the EU. Magnetic materials are also employed in the manufacturing of magnetic sensor devices, which are widely used in many consumer and industrial applications. The current manufacturing process of high-performing magnets heavily relies on the use of rare-earth elements, whose mining and processing produce a remarkable amount of polluting waste. Moreover, EU countries are currently importing 98% of their rare-earth elements requirements, thus facing a serious risk of supply shortage and price volatility. To reduce the  environmental footprint of rare-earth supply and to  mitigate the consequences of their supply shortage,  private and governmental entities in the EU have been fostering the search for alternatives to rare-earth elements in the development of novel magnetic  materials. Established by a consortium comprising both private companies (Robert Bosch GmbH, SIEMENS AG) and academic institutions (Danube University Krems, Max Planck Society, University of Uppsala, Leibnitz Institute for State and Materials Research, Centre National de la Recherche Scientifique), MaMMoS aims at developing an integrated toolchain for the characterization and modeling of magnetic materials from the nanoscale to the magnetic device scale, focusing primarily on improved critical-rare-earth-reduced and rare-earth-free hard magnetic materials in the systems Nd-Ce-Fe-B, Mn-Al and Fe-N. AI and ML will play a central  role in the development of the proposed toolchain,  especially for the integration of experimental and  simulation data. In particular ML models will be  employed to correct the systematic error of ab-initio  simulations, to complement simulation data in  domains not covered by the simulations, and to make predictions of materials properties. Specific feature  engineering and dimensionality reduction techniques  are currently being applied to the training data, with  the purpose of encoding magnetic material properties into a low-dimensional vector space. We will then train ML models of increasing complexity to unveil the hidden patterns in our datasets. The feature encoding map, the best-fit model parameters, the model outcomes, and their limitations will be interpreted and validated iteratively by domain scientists, also  exploiting existing optimization toolkits like Dakota. The final models and all generated data will be  publicly shared open-source, in accordance with the FAIR principles. In conclusion, the MaMMoS project is expected to have a remarkable impact on the materials science community. The final modelling and characterization suite, powered by AI, will be unique in its nature, as has been designed to simulate the behavior and properties of magnetic materials at different length scales, from the atomic to the macroscopic. It can help researchers and engineers design and optimize advanced magnetic materials with new functionality for various critical applications, such as electric motors and magnetic sensors. By using and improving methods of AI, it can also accelerate the material discovery and innovation process by learning from data and building new predictive models. more

The overall goal of Plasma-PEPSC is to enable scientific breakthroughs in plasma science Grand Challenges through exascale computing and extreme-scale data analytics. Specifically, we aim to enable unprecedented simulations on current pre-exascale and future exascale platforms in Europe to control plasma-material interfaces, optimize magnetically confined fusion plasmas, design next-generation plasma accelerators and predict space plasma dynamics in the Earth’s magnetosphere.
We achieve these goals by maximizing the parallel performance and efficiency of four European flagship plasma codes with a large user base: BIT, GENE, PIConGPU, and Vlasiator. Here, we will build on algorithmic advances (regarding load balancing, resilience, and data compression) as well as on programming model and library developments (MPI, accelerator and data movement APIs and runtimes, in-situ data analysis). We ensure an integrated HPC software engineering approach for deploying, verifying, and validating extreme-scale kinetic plasma simulations that can serve as a community standard. We will establish a continuous and integrated co-design methodology to provide/receive direct input to/from the design and development of the EPI Processor and accelerator, will exploit synergies through collaborations with other CoEs, EuroHPC, and Competence Centers for cross-fertilization, adoption and full exploitation of the Plasma-PEPSC codes. Plasma-PEPSC brings together an exceptional, interdisciplinary group of highly-recognized leading scientists from academia, research centres, and HPC centres, with decades of experience in algorithmic and method developments, extreme-scale plasma simulations, and application optimizations with high involvement in strategic EuroHPC projects and initiatives.
MPCDF takes a leading role in Plasma-PEPSC by providing the technical director of the project.

BioExcel is the leading European Centre of Excellence for Computational Biomolecular Research. Established in 2015, the centre has grown into a major research and innovation hub for scientific computing. BioExcel develops some of the most popular applications for modelling and simulations of biomolecular systems. A broad range of additional pre-/post-processing tools are integrated with the core applications within user-friendly workflows and container solutions. The software stack comes with great performance and scalability capabilities for extreme-scale utilization of the worlds largest high-performance computing (HPC) and high-throughput computing (HTC) compute resource. BioExcel has developed an extensive training program to address competence gaps in extreme-scale scientific computing for beginners, advanced users and HPC/HTC system maintainers. The centre maintains an extensive and growing network of industrial researchers in the pharmaceutical, chemical and food industries, and offers tailored products and consultancy services, while code development is done in close collaborations with hardware and software vendors to ensure compatibility and support for cutting-edge features. BioExcel works closely with various governmental, non-profit, educational and policy projects and initiatives.

The main objective of the ADMIRE project is to create an active I/O stack that dynamically adjusts computation and storage requirements through intelligent global coordination, malleability of computation and I/O, and the scheduling of storage resources along all levels of the storage hierarchy. To achieve this, we will develop a software-defined framework based on the principles of scalable monitoring and control, separated control and data paths, and the orchestration of key system components and applications through embedded control points.

DICE aims to enable a European storage and data management infrastructure for EOSC, providing generic services and building blocks to store, find, access and process data in a consistent and persistent way.

MPCDF is a partner in this project that will offer 14 state-of-the-art data management services together with more than 50 PB of storage capacity.

EOSC-hub brings together multiple service providers to create the Hub: a single contact point for European researchers and innovators to discover, access, use and reuse a broad spectrum of resources for advanced data-driven research. For researchers, this will mean a broader access to services supporting their scientific discovery and collaboration across disciplinary and geographical boundaries. The project mobilises providers from the EGI Federation, EUDAT CDI, INDIGO-DataCloud and other major European research infrastructures to deliver a common catalogue of research data, services and software for research. 
EOSC-hub collaborates closely with eInfraCentral, EOSCpilot, GÉANT 4.2, OpenAIRE-Advance and the RDA Europe 4.0 projects to deliver a consistent service offer for research communities across Europe.
EOSC-hub is funded by the European Union’s Horizon 2020 research and innovation programme under grant agreement 777536. The generous EU funding received by the project is complemented with a contribution from the EGI Foundation and its participants, and in-kind contributions made available by service providers of the EGI Federation.

With over 10000 members from 145 countries, the Research Data Alliance (RDA) provides a neutral space where its members can come together to develop and adopt infrastructure that promotes data-sharing and data-driven research

In the context of the collaboration of the MPG with Latin America, the National High Technology Center (CeNAT) of the University of Costa Rica and MPCDF hold a CONARE grant for "Advancing plasma physics computer simulations with the latest high performance computing techniques". The consortium focuses on the assessment and application of modern performance-portability frameworks towards exascale computing.