eCSE programme: funding software development for UK computational science
Posted: 2 Nov 2018 | 23:00
The embedded Computational Science and Engineering (eCSE) programme has allocated funding to the UK computational science community over a period of six years. Integral to ARCHER, the National HPC Service, there has been a series of regular eCSE Calls to fund software development activities.
The last of the Calls has now closed and all funding has been allocated. Although a number of projects are still on-going, this seems a good time to review the benefits of the programme and to see whether its aims have been met.
The programme aimed to:
• Enhance the quality, quantity and range of science produced on the ARCHER service through improved software.
• Develop the computational science skills base, and provide expert assistance embedded within research communities, across the UK.
• Provide an enhanced and sustainable set of HPC software for UK science.
To achieve this we set out to provide a high quality, fair and objective eCSE selection process, delivering maximum value to the community. Selection was made by a series of independent panel members, we had regular Calls, and the programme was not-for-profit, with all funds being spent on projects. In total, 46 institutions were involved across 100 eCSE projects.
A key outcome of the eCSE programme relates to people. We aim to develop the computational science skills base and provide expert assistance and high quality RSE work embedded within research communities across the UK. As the map (right) shows, technical staff have been spread all across the UK.
Enhancing the quality, quantity and range of science produced on the ARCHER service is obviously core to the programme. Scientific output and impact will continue to be delivered throughout the lifetime of the ARCHER service and beyond, hence the full scientific benefit of the programme will not be known and realised for some time. However one metric we can measure is the financial saving achieved from a number of projects. Many projects involved performance optimisation, resulting in a reduction in CPU utilisation and a related financial saving. This saving is re-invested to allow scientists to achieve more science from the same resource allocation.
Carrying out these measurements is tricky, but helps demonstrate the value of investment in software development and research software engineering. Overall the eCSE programme cost around £6m and to date we have seen a reported benefit of around £21m, a more than three times return on investment.
Image shows Hull University’s VOX-FE, a bespoke bone modelling software tool for in silico experiments such as testing bone growth under stressed conditions. The eCSE programme enabled EPCC to work with Richard Fagan at Hull to greatly improve VOX-FE’s modelling capabilities on machines such as ARCHER.
Lorna Smith, EPCC