CBM4scale

On 13 May 2025, the Inno4scale project reached its conclusion with the final event, the Inno4scale Innovation Day, which was hosted at the Universitat Politècnica de Catalunya in Barcelona. The event convened researchers, innovators, and stakeholders from across Europe to reflect on key achievements and discuss the future of high-performance computing (HPC) algorithms as the field transitions into the exascale era. The day commenced with introductory statements from the hosts and the project consortium team,

A novel Mission in Exascale CFD The innovation study STRAUSS, short for “Scalable Task-Parallel Multigrid Solvers” is led by Dr. Niclas Jansson from KTH Royal Institute of Technology and aims to develop highly scalable algorithms to handle the computational challenges posed by exascale systems, particularly for Computational Fluid Dynamics (CFD). Through innovations in parallelism and algorithmic efficiency, the project aims to unlock the potential of European supercomputers like LUMI and Leonardo. Dr. Jansson gave an

The exaSIMPLE project addresses critical challenges in high-performance computing (HPC) and computational fluid dynamics (CFD). Led by Dr. Guilherme Vaz, exaSIMPLE embeds machine learning (ML) directly into CFD algorithms based on the SIMPLE algorithm, which has remained largely unchanged for decades. BlueOASIS, the project coordinator, is leading the development with a focus on CFD, AI, and scientific programming, ensuring smooth project management and coordination. INESC-TEC, a prominent research institute in Portugal, provides invaluable HPC resources,

As computational power moves towards the exascale era, the complexity of scientific simulations continues to increase. A key challenge facing many scientific applications is the efficient and accurate solution of large-scale eigenvalue problems. One such effort to address this challenge is our ISOLV-BSE innovation study. Led by José E. Román from the Universitat Politècnica de València, ISOLV-BSE targets an important class of problems involving structured pseudo-Hermitian matrices in the context of the Bethe-Salpeter Equation (BSE)

As we approach the exascale era in high-performance computing, the need for innovative algorithms that can efficiently handle massive datasets and complex computations is increasing. The CBM4scale innovation study focuses on the development of a novel matrix compression format and associated algorithms to improve the performance of Graph Neural Networks (GNNs) and other scientific applications on exascale systems. In a recent interview, Prof. Dr. Siegfried Benkner, co-principal investigator of CBM4scale and head of the Scientific