Pilot on new algorithms for HPC applications

Proposals are expected to define an effective outreach approach for identifying and attracting innovative ideas to efficiently exploit current and future European exascale HPC systems. A mechanism involving financial support to third parties should be implemented, which will adequately stimulate such innovation potential of researchers and software engineers participating in the action. Third-party grants should be motivated by one of the following approaches. : 

1. Requirement-driven projects, which aim to meet the specific needs of an application community.  
2. Exploration-driven projects, which focus on high-risk/high-gain algorithm development, with the goal of demonstrating a proof-of-concept for a new algorithmic approach.

To achieve the objective of collaboration, an effective monitoring process should be implemented to identify valuable and possibly groundbreaking results. Proposals should define a clear approach for facilitating productive interactions between relevant stakeholders - including algorithm developers (third parties receiving financial support), HPC application developers and users to maximize the uptake, adaptation, and integration of the proofs-of-concept delivered by the action. Proposals should allocate sufficient resources for collaboration activities. The collaboration tasks and pre-allocated resources will be finally defined by the consortium after the evaluation together with EuroHPC JU and in coordination with other relevant grants. 

Key parameters of mechanism involving financial support to third parties (FSTP):

Each third-party which received financial support in this action will deliver by the end of the project at least: 

1. A proof-of-concept implementation running on real hardware including meaningful benchmark data that demonstrates significant supremacy compared to existing solutions. 
2. With respect to the base line provided in the application for financial support: a revised estimate of the reduction for time-to-solution or energy-to-solution of a list of applications and use cases, based on the best available benchmark data. 
3. An impact assessment on the reduction of HPC resource consumption considering the present usage pattern and potential use cases of the algorithms. 
4. Documentation of the new algorithms and their technical implementation including relevant test cases to prove correctness and ensure reproducibility of the results. 

Each third-party will provide the specified deliverables in accordance with the documented instructions and templates provided by the consortium, ensuring they meet all defined conditions and quality standards. The consortium will then ensure these deliverables will be available in a public repository including a description, all relevant data, code and documentation for easy use.

The procedure for disbursing financial support will be implemented through open and competitive calls widely disseminated among the EU and Participating States of the EuroHPC JU, based on predefined evaluation criteria encompassing:  

1. Innovation potential: Does the proposed idea have the potential to drive scientific or industrial innovation and/or address significant societal challenges? Are the expected outcomes innovative or clearly distinctive compared to existing solutions (fundamentally new and innovative HPC algorithm are sought - porting, reimplementation, incremental improvements or parallelization of an existing implementation are not sufficient).  
2. Approach and methodology: Are the proposed activities and planning appropriate and effective to exploit the capabilities of the current and upcoming European HPC infrastructure in the most efficient way.

The process should also aim at reducing administrative burden for third parties, while at the same time ensuring sound financial management. In particular, the application forms and procedures should be tailor-made and appropriate to the technical and managerial capacities of the targeted applicants and scope of the projects. Calls for proposals must remain open for at least two months. The call documents should be agreed with the Granting Authority before the launch of the call. 

Beneficiaries must ensure transparency with adequate publication of calls for proposals and prevent conflicts of interest throughout the entire award procedure. They will have to clearly demonstrate this in their application and report on it throughout the project. 

The consortium must publish the outcome of the call(s) on their website, including a description of the selected projects, award dates, project durations, final recipients’ legal names and countries of establishment. The indicative timeline for this publication is within two months after the submission deadline of the call the proposal or equivalent date for open-end calls. 

The financial support should cover personnel costs for experienced researchers (post-doctoral) whose roles and qualifications should be demonstrated in the application for FSTP. R&D activities should focus on the development of novel HPC algorithms and costs associated with non-R&D activities - such as consulting, project management, administrative support, or dissemination – are, in principle, not eligible.