Advancing the transition towards exascale capabilities by developing or scaling up existing parallel codes, resulting into effective applications to solve scientific, industrial or societal challenges and addressing the needs of the user communities.
The Centres of Excellence (CoE) address scientific and industrial applications and user communities that run world leading codes or ensembles that aggregate to pre-exascale workloads, and may require exascale resources in specific cases.
All CoEs should be user-driven and inherently committed to co-design activities to ensure that future HPC architectures are well suited for the applications and their users (both from academia and industry), providing a high performance and scalable application base.
CoEs should federate existing resources around Europe, exploiting available competences, and ensuring multidisciplinary (combining application domain and HPC system, software and algorithm expertise) and synergies with national/regional programmes.
CoEs should further enlarge and expand these capabilities all over Europe, in particular by including users communities from EU countries currently developing and advancing their HPC infrastructure and ecosystem (e.g. countries not currently hosting Tier 0 and/or Tier 1 supercomputing facilities).
Requirements for CoEs:
- The provision of services supporting different usage models for the community needs, including developing, maintaining, optimising (if needed re-designing) and scaling HPC application codes, addressing the full scientific/industrial workflow, particularly covering data aspects; testing and validating codes and quality assurance,
- Enhancing HPC applications and ensembles of coupled applications towards highly scalable, optimised codes.
- Streamline development, collaboration, automated testing and deployment processes throughout the application development and maintenance cycle, for example, by provisioning and using state-of-the-art development tools, platforms and software management models.
- Fostering pre-exascale-oriented codes and innovative algorithms that address societal challenges and/or are important for key scientific and industrial applications. Demonstration of clear societal, industrial, and/or scientific benefit is mandatory.
- Provision of codes aimed at capacity pre-exascale systems for ensemble workloads that might not need tight interconnection between runs. Proposals must convincingly demonstrate the ensemble pre-exascale capabilities and needs.
- Collaboration with EuroHPC JU systems and infrastructure federation initiatives to implement a robust and reliable automated deployment process for applications, in order to make novel developments timely available to the EuroHPC JU user communities.
- Involvement in co-design activities (hardware, software, codes), including the collaboration with HPC vendors and the identification of suitable applications relevant to the development of European HPC technologies towards exascale.
- Enlarging and expanding HPC applications development and use, in particular for new user communities in EU countries and countries associated to Horizon Europe that are members of the EuroHPC Joint Undertaking currently developing and advancing their HPC infrastructure and ecosystem.
- Ensure wider access to codes and foster their uptake by user communities, in particular industry and SMEs.
- Federating capabilities and integrating communities around exascale computing in Europe.
- Address the skills gap in the targeted domain by specialised training and capacity building measures to develop the human capital resources for increased adoption of advanced HPC in industry and academia.
- Complementarity and synergy with National HPC Competence Centres and EU projects.
- Include clear KPIs on the optimal employment of current and/or emerging HPC technologies, allowing the assessment of the progress towards the objectives, both in terms of outputs and ultimate impact.
Proposals should ensure the cooperation with complementary projects launched by the EuroHPC JU including also the need to establish from the beginning of this cooperation appropriate IP exploitation agreements and should provide preliminary benchmarking data on new and emerging HPC technologies. Proposals should also clearly demonstrate that all partners in the consortium have a significant and justified role, including appropriate deliverables under their responsibility which cover their specific contributions.
06 April 2022