Call on Centres Of Excellence For Exascale HPC Applications

Deadline: Jan 9, 2024  
CALL EXPIRED

Centres of Excellence advancing specific Lighthouse Exascale Applications, at the frontier of technology and relevant for the communities of HPC users, that enable and promote the use of upcoming exascale and post exascale computing capabilities in collaboration with other High Performance Computer (HPC) stakeholders. They should implement concrete actions to increase the performance of applications and exploit these advanced computing capabilities. The goal is to develop or scale up existing application codes towards exascale performance, resulting into tangible benefits mainly for scientific challenges. Proposals for Centres of Excellence - Exascale Lighthouse applications will exploit existing federated resources around Europe, developing available competences, and ensuring multidisciplinary (combining application domain and HPC system, software and algorithm expertise).

This topic builds and complements the HORIZON-EUROHPC-JU-2021-COE-01-01: Centres of Excellence preparing applications in the Exascale era call.

Proposals should focus on the development of specific and clearly identified applications and codes, convincingly demonstrate their exascale capabilities and needs, and present a detailed software development plan with clear timeline for the implementation including quantitative KPIs, milestones and deliverables demonstrating the achieved improvements in the strong and weak scaling of each code with respect to the baseline and different system architectures. This includes codes and tools that support the analysis and assessment of academic or industrial applications with potential for performance optimisation that can exploit the current and future advanced computing capabilities. The software development plan, covering the identified applications and codes, should be central to the proposed work and most resources should be allocated to these activities. Research activities on the basis of use cases are not within the scope of the action and use cases should be limited to test runs required for development purposes such as regression tests.

Proposals for Centres of Excellence in Topic HORIZON-EUROHPC-JU- 2023-COE-03-01 must clearly identify one of the following the Exascale Lighthouse application areas:

1. Personalised Medicine/ Digital twin of the human body
2. Human Brain research & neurological disorders

Only one proposal will be selected per Exascale Lighthouse applications topic identified above. Proposals should also be able to articulate clearly the scientific grand challenge(s) which will be addressed by the applications and why the exascale performance is needed.

Targeted applications should be relevant for communities of HPC users as well as for future EuroHPC JU systems to be acquired. Proposals should be inherently committed to co-design activities to ensure that future HPC architectures are well suited for the applications and their users.

Requirements for CoEs:

• Clear identification of the targeted applications and all related codes that will be developed or used in the proposal, including their user basis and the global impact in their domain. The ownership and license of each code must be listed in the proposal. Only applications (software) which are owned or controlled^[1] by the consortium members are eligible. In addition, proposals should provide lists of all relevant codes and use cases. The JU provides a specific template for this purpose.
• Present a clear baseline to demonstrate the exascale suitability for all key applications, use cases and workflows considered in the proposal substantiated by corresponding strong and weak scaling plots. Regular deliverables with updated figures should be provided according to the software development plan and at least every six months.
• Describe the European user communities of the targeted applications, the current and predicted use on EuroHPC infrastructure for the indicated codes as well as the impact of the planned developments on the European users.
• Demonstrable advances of the targeted HPC applications towards highly scalable, optimised flagship codes and exascale performance (both computing and extreme data). This includes developing, maintaining, porting, 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. This also includes horizontal tools and services that can be applied to parallel codes in any application domain to analyse and improve their performance.
• Addressing the exascale and post exascale related technical challenges, such as load balancing; resilience; heterogeneity programming models, in particular accelerator-based architecture programming; run-time systems; workflow management tools; development environments and production environments and how they are tackled.
• 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 and collaboration with European initiatives (e.g. EPI, RISC- V, EuroHPC JU Pilots). This includes the expected impact of the applications on emerging HPC technologies and how the development of hardware is influencing the software development.
• Activities to improve the energy efficiency of applications, algorithms, methods, libraries and/or tools.
• 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.
• Federating capabilities and integrating communities around exascale computing in Europe.
• 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.
• Coordinate within the European ecosystem, including Competence Centres, to address the skills gap in the targeted exascale applications and codes, by specialised training and capacity building measures to develop the human capital resources for increased adoption of exascale solutions.
• Coordinate with Competence Centres to ensure wider access to codes and foster their uptake by scientific user communities.
• Proposals should ensure the cooperation with complementary projects launched specifically in the area of the “EuroHPC-2020-01-a: Advanced Pilots towards the European Supercomputers” 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.

In addition, proposals should ensure collaboration with other Centres of Excellence for HPC applications, and other national and EU funded activities that focus on similar or complementary objectives for HPC codes and applications, in order to maximise the synergies and optimise such codes and applications for current and future architectures of EuroHPC supercomputers. This includes participation in the common continuous integration and deployment platform developed by Centres of Excellence for HPC applications selected in call HORIZON-EUROHPC-JU-2021-COE-01 and HORIZON-EUROHPC-JU-2023-COE-01 and the associated Coordination and Support Action CASTIEL 2. Selected proposals are expected to accede the collaboration agreement between existing Centres of Excellence and CASTIEL 2. 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 the specific contributions of each partner. Moreover, applications should include a well-balanced consortium with an appropriate distribution of resources and responsibilities, in line with the proposed work plan. Consortium members are expected to contribute to the project with a share of at least 5% of the total declared personnel resources.

The expected duration of this action is 30 months with a foreseen project start on 01/07/2024.

[1]This may include software owned by third parties which is provided under a permissive license. In such a case the consortium must demonstrate in the proposal the ability to develop the software independently of the owner (for example, demonstrate sufficient knowledge of critical software components).