Synopsis
The ExCALIBUR programme is built around four fundamental pillars that describe the principles that guide the development of research under ExCALIBUR and are designed to ensure that the outcomes are future-proofed against the constantly evolving landscape of hardware design. The four pillars are:
- separation of concerns: separating the mathematical problem from the computer science implementation
- co-design: holistic, collaborative system design by mathematicians, domain scientists and computer scientists
- data science: new workflows to manage and analyse vast volumes of simulation data
- investing in people: interdisciplinary Research Software Engineer (RSE) career development driven by forward-looking scientific software design.
Expectations of DDWGs
The first opportunity produced the currently funded DDWGs. The applications should demonstrate plans for future work and how these have been informed by the work of the DDWGs towards the stated expectations.
DDWGs are expected to conduct a mixture of simulation code design and development, and community building activities that will engage relevant computational and user communities.
DDWGs are expected to deliver the following outputs:
- a strategic research agenda that clearly articulates the research challenges to be overcome, opportunities, key risks and mitigations, and sets out a detailed approach to addressing these to enable development of exascale-ready software by the mid-2020s
- evidence that the proposed approach has been developed collaboratively with potential beneficiaries, including co-design with industry where appropriate
- demonstration of the feasibility of the proposed approach through proof-of concept studies and research outputs. Scaling should already be demonstrated at the petascale and relevant software development should already be underway.
Characteristics of high-priority use case
A high-priority use case for exascale software development is defined by the ExCALIBUR programme and accepted by the ExCALIBUR programme board and steering committee as: a co-ordinated range of activities, which aims to develop simulation code with a focus on an application or applications pre-identified by the relevant communities as benefitting from exascale software development.
A high priority use case for exascale software development has the following characteristics:
- it provides a step-change in simulation performance and provides solutions that are not currently feasible, consistent with the enhanced performance of exascale computing
- it enables high-quality, high-impact research in multiple areas of strategic importance
- it produces applicable and scalable solutions that can be applied across a range of architectures, including non-exascale systems
- it provides a national and international focal point for the relevant research communities, including the development of partnerships with complementary initiatives in the UK and internationally.
Scope
The ExCALIBUR programme aims to encompass all fields of research within UKRI that can benefit from or advance the development of the UK’s exascale software space.
Changes in computer architectures and how we use them can cause directional shifts both in the technology and the directions of human inquiry that rely on it. These technological advances, which are themselves evolving, coupled with algorithm and software development, will offer an enormous opportunity to address questions which until now have been beyond our reach.
Examples of the resulting breakthroughs in research areas where the UK is world-leading include:
- expanding the frontier of fundamental sciences
- climate, weather and earth sciences
- computational biology
- computational biomedicine
- engineering and materials
- digital humanities and social sciences
- mathematics and science of computation.
The ExCALIBUR programme partners, the Met Office and UKAEA have already funded use case projects in fusion modelling, and climate, weather and prediction. This opportunity will look to fund use cases which encapsulate some of the remaining research areas.
High priority use case aims
This opportunity is aimed towards applicants with the capacity to produce a developed proposal which incorporates the four ExCALIBUR pillars and delivers the following:
- a high priority use case for exascale software
- active knowledge dissemination strategy for the communities that the use case serves
- active collaboration, outreach and engagement across the ExCALIBUR programme including future funded activities and groups.
A use case should enable both strategically important research and provide lessons that can be applied to other codes and fields, maximising the impact across the scientific modelling and simulations communities.
Although closed to the DDWGs, it is expected that these groups are dynamic and will have evolved over time (for example, new entrants to the group or merging of the groups) therefore changes to the teams are acceptable.
Collaborations with industry and international groups are encouraged. Applicants should engage with the ExCALIBUR Hardware and Enabling Software Group or the EPSRC team for information on facilities which are available.
Further funding for a high-priority software use case
The DDWGs were funded to identify a high-priority use case and develop the foundations for this use case within the relevant communities that would underpin exascale software development in the UK.
The high-priority use case is a core tranche within ExCALIBUR, and funded grants are expected to show:
- they utilise the lessons learnt from the current DDWGs to produce and develop exascale-ready codes and software relevant to the use case, considering the maintenance and sustainability of this code
- progress has been made in identifying and addressing a high-priority use case built upon the work of the DDWGs
- they establish two-way knowledge exchange with the wider research community, industry and internationally for your use case on the behalf of the programme
- proposals include consideration of how training and upskilling of researchers, RSEs or industry is supported to ensure they are prepared for the potential of exascale software and architecture
- how they will integrate and optimise their development of scientific code in parallel with the undertakings of the ExCALIBUR cross-cutting projects and other activities to be funded by the programme.
Knowledge exchange coordinator
Knowledge exchange (KE) is a vital component of achieving the objectives of the ExCALIBUR programme. It will ensure integration across the programme activities where researchers are developing software and algorithms in preparation for future exascale systems. Additionally, connections are required with potential beneficiaries in academia, public sector research establishments (PSREs) and industry to contribute to these designs and the dissemination of outcomes.
Therefore, proposals must include a named co-investigator or research co-investigator who will have the role of a knowledge exchange coordinator to lead these endeavours. Flexible funds can be requested to deliver activities to aid knowledge exchange and will be managed by the principal investigator and KE coordinator. Expectations for this role include, but are not limited to:
- identify opportunities for knowledge exchange within their project, with other ExCALIBUR programme projects, and with other relevant national and international projects
- identify opportunities for knowledge exchange to develop and maintain a two-way flow of engagement and dissemination with industry and relevant national and international research communities
- develop a plan to increase the awareness of the proposed activity and the ExCALIBUR programme. The plan should include a rationale of the flexible funds requested to support this and timescales to accomplish this
- collaborate with other ExCALIBUR knowledge exchange coordinators as a network to deliver the programme’s knowledge dissemination strategy.
Funding available
Up to £8 million is available to fund up to three projects for a duration of three years at 80% full economic cost. Financial profiles must be confirmed with EPSRC before grants can begin. Costs should include:
- flexible funds for knowledge exchange activities (expected to be equivalent to 10% of the total funds requested)
- researcher time
- consumables
- technical staff time
- support staff.
Equipment over £10,000 in value (including VAT) is not available through this opportunity. Smaller items of equipment (individually under £10,000) should be in the Directly Incurred – Other Costs heading.
For more information on equipment funding, please see EPSRC’s equipment page.
