A history of collaboration
MaX Centre of Excellence stands at the forefront of European and global HPC materials research. Since its launch in 2015, MaX has accelerated discovery and innovation in materials science by developing and adapting its lighthouse codes for upcoming exascale and post-exascale computing systems.
Built on strong European collaborations, shared strategies, and deep integration within the HPC ecosystem, MaX has become a reference point for computational materials research in Europe and beyond. These partnerships continue to drive progress and strengthen Europe’s position in advanced materials design.
A decade of HPC and materials excellence
Now in its third phase, the MaX Centre of Excellence brings together 14 partners and 2 affiliated entities from six European countries including Italy, France, Spain, Slovenia, the Czech Republic, and Germany. Many have contributed since the project’s early phases in 2015 and 2018, ensuring continuity and expertise in both code development and industrial collaboration.
Over nearly a decade, MaX partnerships have aligned with European and EuroHPC policies, shaping a powerful network for quantum materials simulation and HPC innovation. These collaborations have made MaX a cornerstone of the European HPC and materials research ecosystem.
Below, we highlight the collaborations that have accompanied MaX since 2015 and the previous funding periods that have helped MaX become a cornerstone of the European application and HPC ecosystem.
[2018-2021] Second funding period
MaX Materials design at the eXascale. European Centre of Excellence in materials modelling, simulations, and design.
Objectives
The main goal of MaX2 was to allow the pre-exascale and exascale computing technology expected in Europe in the 2020’s to meet the demands of a large and growing base of researchers committed to materials discovery and design.
Achievements
We designed an innovative software model, based on the concept of separation of concerns, that improves the performance of MaX community codes on heterogeneous hardware architectures.
We built an integrated ecosystem for the convergence of high performance and high throughput computing. The integrated ecosystem allows steering the hundreds of millions of simulations that are needed to optimise the properties and performances of a material or a device, with robust and reproducible workflows.
We co-designed and co-developed hardware and software suitable for the exascale transition by simultaneously working on architectures, computing constraints, and scientific goals.
We organized innovative professional and educational trainings to ease the access to MaX lighthouse codes, engage academic and industrial communities, and consolidate a pool of well trained users and developers of materials science applications.
[2015-2018] First funding period
MaX Materials design at the eXascale.
Objectives
The main goals of MaX1 were to develop, maintain, and optimize community codes for materials science simulation. Particular attention was put on improving their capabilities and reliability, ensuring provenance, preservation, and sharing of data and workflows. MaX1 also provided hardware support to help transition to the exascale architectures, as well as other related services offered to the MaX community.
Achievements
We developed a programming framework for materials simulations that separates quantum computing engines from specialized, low-level domain-specific libraries. This separation enables advanced features to be more easily integrated with and used by a variety of quantum engines, making it easier to extend the capabilities of materials simulations and broadening their potential applications.
We expanded code capabilities to better design materials and functions of relevance for industrial and societal needs.
We developed an informatics ecosystem to automate high-throughput calculations and automatically store the resulting data in graph databases. This system provides a ready-to-use environment that ensures turnkey solutions for managing workflows and data sharing and provenance. It also supports reliable data storage, preservation, reproducibility, and reuse.
To prepare for the exascale transition, we evaluated advanced programming models, novel algorithms, domain-specific libraries, in-memory data management, software/hardware co-design and technology transfer actions with other FET-HPC initiatives in the H2020 program.
We supported the MaX communities (e.g., computational and materials scientists, software scientists, software vendors, end-users in industry and in academic research) by providing different consulting services, including specialized user support forums on each reference code.
We trained end-users and software developers via schools and workshops. We supported university programs via pilot courses and developed a MOOC model for training schools.
