Simula’s HPC department awarded funding from new EU initiative
Funding from EuroHPC JU is highly competitive with proposals measured over three criteria; excellence, impact, and quality and efficiency of the implementation. Proposals can get a maximum score of 15 points. Led by Professor Xing Cai, head of the HPC department, Simula is a key partner in two proposals “SparCity” and “MICROCARD” that received 15 and 14.5 points, respectively – which ranked them number 1 in their respective categories.
EuroHPC was launched in 2018 to enable EU and participating countries to coordinate their efforts and share resources to make Europe a world-class supercomputing infrastructure.
“Now is an exciting time with rapid developments of hardware and software technologies,” says Professor Xing Cai, head of the HPC-department at Simula. “It is thus a great opportunity for researchers at Simula to collaborate with internationally leading experts, through these two EuroHPC projects, to bring innovative use of HPC to real-world applications.”
The two projects, “SparCity” and “MICROCARD” are part of 19 other projects that will contribute to EuroHPCs objective for the EU to become a world leader in supercomputing. The projects will run for a period of 3 and 3.5 years, respectively.
There are immense volumes of data that need to be processed and studied, and these datasets are growing in both size and complexity. One such complexity is sparse data, that is, data sets where one can dramatically reduce computing time and save storage space by exploiting the intrinsic structure of the data. Computations on sparse data structures are already notoriously difficult for existing HPC systems to achieve good efficiency, but the rapidly evolving hardware landscape will make the challenge even more pronounced.
The overall objective for SparCity is to create a supercomputing framework of efficient algorithms and coherent tools for maximizing the performance and energy efficiency of sparse computations on emerging HPC systems, while also opening up new usage areas for sparse computations in data analytics and deep learning.
Realistic simulations of cardiac electrophysiology in structurally abnormal tissue require resolving the individual cells and their interconnections, which cannot be handled by the state-of-the-art mathematical models, numerical strategies, simulation software. Innovations in all these aspects are thus urgently needed to investigate the detailed physiological mechanisms of many heart diseases, to develop more effective treatments and drugs.
The overall objective for MICROCARD is to develop a production-ready simulation platform for cardiac electrophysiology on models with micrometre resolution. This platform will be designed to work on future exascale computers, by adopting the latest developments in HPC methods, as well as problem-tailored algorithms, to apply on all the levels of future-embracing hardware platforms.
Read more about the EuroHPC here: www.eurohpc.eu