In engineering and physical sciences research, a world-class infrastructure comprises not just large high-specification equipment and access to central facilities but also the distributed investment in laboratory-based equipment and the technical support needed to ensure its productivity.
In engineering and physical sciences research, a world-class infrastructure comprises not just large high-specification equipment and access to central facilities, but also the distributed investment in laboratory-based equipment and the technical support needed to ensure its productivity.
Use of computational approaches such as modelling and simulation, software development and data analysis capabilities are a pervasive component of 21st century research infrastructure, and are critical to sharpening the UK’s industrial competitiveness.
Through our future-proofing state of the art research infrastructure delivery plan priority we acknowledge the relationship between laboratory-based, distributed facilities and large-scale campus-based facilities. We also seek to ensure maximum availability and sharing of equipment across organisations, to realise the benefits from our infrastructure investments.
The Research Infrastructure theme focuses on ensuring the right infrastructure is available to EPSRC-sponsored researchers at the right time, in the most economic way possible.
Research infrastructure relationships
Visualising our portfolio (VoP) is a tool for users to visually interact with the EPSRC portfolio and data relationships. Find out more about research theme connections and funding for research infrastructure.
EPSRC considers e-infrastructure to be computing facilities and software for the support of research. Within e-infrastructure, our support is concentrated around eScience, software, high performance computing and cloud.
The EPSRC e-infrastructure team, in collaboration with both its strategic advisory team (SAT) and key members of the community, have developed an e-infrastructure strategy and software infrastructure strategy to enable world-leading research both now and in the future.
A coherent strategy for developing and delivering the UK’s future e‐infrastructure needs is essential in driving forward the continued development of a globally competitive research base within the UK.
The EPSRC, along with councils under UKRI, the funding councils, Innovate UK and the Department for Business, Energy and Industrial Strategy, plays a key role in developing the strategy as well as delivering the funding to support e‐infrastructure in the UK.
The development of a sustainable and cutting edge e‐infrastructure ecosystem is vital in allowing EPSRC to deliver its strategic goals and support excellent and innovative science and engineering research. EPSRC is pleased to welcome Professor Mark Parsons as Special Strategic Advisor on e-infrastructure. This additional advice will ensure that the needs of the EPSRC community are identified and considered in the development of the UKRI e-infrastructure strategy.
The EPSRC has formulated the EPSRC e-infrastructure strategy 2018 and EPSRC software infrastructure strategy 2018 with the help of key stakeholders. Both documents will be updated on a regular basis.
EPSRC e-infrastructure strategy 2018 was developed with UKRI to deliver the UK’s future e-infrastructure needs and is essential in driving forward the continued development of a globally competitive research base within the UK.
EPSRC software infrastructure strategy 2018 is a strategy for investing in software, developed in conjunction with the community and designed to ensure that our current and future funding adds value to the complex and evolving e-infrastructure ecosystem.
Computational resources for COVID-19 research
UKRI supports a wide range of computational capabilities that can be requested by researchers seeking to contribute to the understanding of, and response to, the COVID-19 pandemic and its impacts.
High performance computing
The coordination of high performance computing (HPC) activities for academic research in the UK is the joint responsibility of the UK research councils. This includes the strategic direction of HPC developments in the UK, access arrangements to high performance computing and support services, planning the procurement and location of HPC services, and promoting their widespread use.
Partnership for Advanced Computing in Europe (PRACE)
The UK is signed up to the Partnership for Advanced Computing in Europe (PRACE). PRACE is an initiative to allow researchers to access a world-class suite of supercomputers across Europe, via a central peer review system.
By signing up to PRACE, UK researchers will now have access to different types of first class supercomputer architectures, enabling them to expand and enhance their research capability and productivity. It is also possible to gain access to Tier-1 systems through a process managed by DECI, which runs in parallel to the Tier-0 calls.
Access to PRACE machines
All access to PRACE machines will be managed through a central peer review system.
Calls for proposals are released about every six months and award time on the machines through a variety of different mechanisms.
Peer review will only allocate time on the machines and researchers will not be able to apply for financial support.
More information on the PRACE peer review system can be found on the PRACE website.
Guidance on applying to PRACE
The UK (via EPSRC, STFC and BBSRC) is now paying around £1 million a year to be members of PRACE and to get access to the PRACE machines for UK scientists.
In order to maximise the amount of time obtained, and to help those thinking of applying, the following guidance has been developed by Sylvain Laizet and Debora Sijacki, who have had a lot of experience and success in applying to PRACE. Thanks to them for their help. If anyone else would like to add anything based on their experiences, please contact the ARCHER team at email@example.com.
First of all, it may be obvious but it is very important to read all the information on the PRACE website. The main sources of information are the user documentation pages and the best practice guides on the PRACE website. Make yourself aware of how to apply and what the application procedures are. You might have to apply several times before being successful. The good thing is that you will be able to use the feedback from the referees to improve your application.
You need to see a PRACE proposal in a similar way as a full UK Research and Innovation proposal. Do not wait until the last minute to write your proposal. Make sure that your colleagues read it and give you valuable feedback. The scientific aspect is of crucial importance. You will need to demonstrate novelty, impact and timeliness. This is the most important part of the proposal.
Building a good team with EU partners is essential – each member of the team needs to have a specific expertise. Your experience of using HPC resources in the past and how you will manage using a Tier-0 system are also very important. It is worth emphasising if the simulation data have a legacy value: what will be shared with the scientific community and how.
Aim for big, but it is important to check how much resources are available on a given system (it might be risky to ask for greater than 40% to 50% of the system’s core hours). You must aim for simulations that can only be performed on Tier-0 systems, not on Tier-1 systems.
You will need to produce scalability plots and representative benchmarks. If you do not have representative benchmarks and data to justify the resource request, you need to apply for a preparatory project, which allows you to port your code, to test it, and to collect all the information necessary for a full proposal.
It is important to have a detailed plan of when and how the simulations will be set up, run, data copied and stored and what are the RAM/I/O/storage requirements at any given stage. Also have a detailed project management plan: who will do what in the collaboration, how data will be shared, meeting or teleconference schedule, plans for paper writing, project web page, and so on.
Information on European e-infrastructure projects:
- e-Infrastructure Reflection Group (e-IRG)
- European Grid Initiative (EGI)
- Partnership for Advanced Computing in Europe (PRACE).
The INCITE program – US Department of Energy
Since 1974, DOE’s Office of Science, the nation’s single largest supporter of basic research in the physical sciences, has provided supercomputing resources for unclassified research through the National Energy Research Scientific Computing Center (NERSC) at Lawrence Berkeley National Laboratory.
There are two Leadership Computing facilities at Oak Ridge National Laboratory and at Argonne National Laboratory which provide over one billion processor hours to the INCITE program.
UK researchers can apply directly to the programme.
EPSRC’s investment in software covers the spectrum from new algorithm development at the leading edge of research applications through software development to code maintenance. It also includes training, and community support activities, such as networking. The current portfolio of activities has supported a thriving community of computational scientists who are recognised internationally.
Software as an infrastructure
The large suite of codes used across all areas of research needs to be regarded as a research infrastructure in its own right, requiring support and maintenance along the innovation chain and throughout its life cycle.
EPSRC, with the assistance of the community, has developed a strategy for investing in software, ensuring that our current and future funding adds value to the complex and evolving e-infrastructure eco-system, and supports the needs and requirements of the engineering and physical sciences community.
Software as an infrastructure details the strategic framework for software, which has been developed following input from the UK community.
Software as an infrastructure: action plan sets out the activities that will be pursued as a consequence of the strategy that has been developed.
Software for the future: workshop report – this workshop aimed to map the current UK activities in software and determine the needs for support for software in the future.
You can find out more about: