MaX (Materials design at the eXascale)
This article was accepted into the corpus but its outbound wikilinks were never NER-processed — typical at the deepest BFS hop or when the run's entity cap was reached. No expansion funnel to show.
| MaX (Materials design at the eXascale) | |
|---|---|
.png) | |
| Name | MaX (Materials design at the eXascale) |
| Abbreviation | MaX |
| Established | 2015 |
| Headquarters | Istituto Officina dei Materiali |
| Region | Europe |
MaX (Materials design at the eXascale) is a European project and research community focused on advancing materials modeling, simulation, and design by exploiting exascale high-performance computing. It brings together national laboratories, universities, and supercomputing centers to develop scalable software, training, and workflows for electronic structure, quantum materials, and multiscale modeling. MaX integrates expertise from computational physics, materials science, and computer science to prepare codes and methodologies for next-generation architectures.
Overview
MaX organizes collaborative activities across research nodes and centers such as CINECA, Jülich Research Centre, Barcelona Supercomputing Center, Istituto Nazionale di Fisica Nucleare, and École Polytechnique. The project connects developers of flagship codes with the operators of systems like Leonardo (supercomputer), Summit (supercomputer), Fugaku, and future Exascale computing project resources. MaX aims to align community-driven software stacks with procurement and deployment strategies of facilities including European High Performance Computing Joint Undertaking and national infrastructures like PRACE. Participants include researchers affiliated with institutions such as University of Cambridge, University of Oxford, École Normale Supérieure, Universidad Autónoma de Madrid, and Instituto de Ciencia de Materiales de Madrid.
History and development
MaX was launched in the mid-2010s amid calls from bodies like Horizon 2020 and the European Commission to coordinate computational materials research. Initial phases involved coordination among consortia that included CNR, CNRS, ICN2, and Max Planck Society groups. Subsequent iterations expanded partnerships to cover centers formerly associated with projects such as PRACE, EUDAT, and collaborations with hardware vendors like Intel Corporation, AMD, NVIDIA. MaX’s development paralleled milestones from events including the International Conference for High Performance Computing, Networking, Storage and Analysis and influenced community positions at meetings like ISC High Performance. Over time MaX produced roadmaps that intersect with strategies from EuroHPC, European Research Council, and national funding agencies.
Objectives and scientific scope
MaX targets objectives that include porting and optimizing electronic structure codes, accelerating many-body methods, and enabling high-throughput materials discovery workflows. Scientific scope spans density functional theory as implemented in packages developed by teams from SISSA, University of Trieste, and CNRS; many-body perturbation theory used by groups from University of Rome Tor Vergata and Ecole Polytechnique Fédérale de Lausanne; and quantum Monte Carlo approaches practiced at Caltech and Princeton University. MaX supports research on topics central to projects at CERN-adjacent computational efforts, and integrates with experimental communities such as those at ESRF, European XFEL, and Diamond Light Source. Objectives also include workforce training through schools modeled after initiatives at Argonne National Laboratory and Lawrence Berkeley National Laboratory.
Infrastructure and computational resources
MaX leverages national and pan-European supercomputers and data centers including CINECA Marconi, Jülich’s JURECA, and BSC’s MareNostrum. Resource strategies coordinate allocations via PRACE and EuroHPC JU mechanisms and intersect with procurement by entities like Deutsches Elektronen-Synchrotron and STFC. MaX addresses challenges of heterogeneous architectures introduced by vendors such as IBM, HPE, and Lenovo and integrates accelerators from NVIDIA and AMD Instinct. Data management and FAIR practices draw on standards from EOSC and services inspired by Zenodo and DataCite.
Software ecosystem and flagship codes
MaX supports an ecosystem of community codes and libraries including plane-wave and localized-basis electronic structure packages developed by groups at Università di Torino, Universidad Complutense de Madrid, and University of Barcelona. Flagship codes associated with MaX efforts include widely used projects stemming from collaborations involving SISSA, ICN2, Max Planck Institute for Solid State Research, and University of Vienna teams. The project emphasizes interoperability with libraries and tools such as MPI implementations from Open MPI Project and MPICH, performance tools from Intel VTune and TAU Performance System, and build systems used by CMake. MaX promotes adoption of standards from OpenMP Architecture Review Board and Kokkos-style abstractions to ensure portability across CUDA and vendor-neutral backends.
Collaborations and partnerships
MaX coordinates with European research infrastructures and academic partners including University of Milano-Bicocca, Universidad de Zaragoza, Technische Universität München, and ETH Zurich. Industrial and vendor partnerships involve entities like Siemens, BASF, and technology providers such as Arm Ltd. and Cray Inc. deepening ties with national laboratories such as Oak Ridge National Laboratory and Lawrence Livermore National Laboratory. Collaborative training and dissemination have taken place at conferences such as NeurIPS-adjacent workshops, domain-specific meetings like Materials Research Society symposia, and through links to initiatives at European Materials Modelling Council.
Impact and notable results
MaX has delivered scalable implementations, performance improvements, and community training that accelerated calculations relevant to photovoltaics, catalysis, and quantum materials studied by teams at MIT, Harvard University, University of California, Berkeley, and University of Illinois Urbana-Champaign. Outcomes include optimized kernels enabling larger system sizes on systems like Fugaku and Summit (supercomputer), contributions to open-source releases led by groups at SISSA and ICN2, and published results presented at venues such as APS March Meeting and European Physical Society Conference. MaX’s influence extends to regional HPC strategies, software sustainability practices adopted by centers like CINECA, and capacity building that fed into grant-funded projects via Horizon Europe and national research programs.