This talk will be delivered by IBM and STFC Hartree Centre scientists collaborating through the Hartree National Centre for Digital Innovation, a programme enabling UK companies to take advantage of artificial intelligence, quantum and high performance computing..

Specifically, in this talk we outline the trends in high performance computing through the lenses of numerical frameworks we have been developing. The topics will span advanced discretisation, fast solutions methods and parallel scaling. We have placed particular emphasis on performance portability and flexibility to support a diverse set of use-cases.

To demonstrate this several case studies of modelling complex multi-physics, multi-scale phenomena will be presented. These will range from aerospace engineering design, automotive to extreme physics of nuclear fusion. We will also discuss how finite element methods can enable efficient and scalable simulations of industrial fluid flow problems. In particular, we will show how this is achieved in PyFR, a cross-platform coupling of Python to a domain-specific language, and discuss the recent improvements made to support recent work at the exascale.

In this webinar you will discover:

• An introduction to the HNCDI partners – STFC Hartree Centre and IBM Research
• How we have used modern software frameworks, such as AMReX and PyFR, to create fast, scalable and adaptable solvers for sets of partial differential equations.
• High-performance computing case studies covering aerospace and automotive engineering design to extreme physics of nuclear fusion.
• Funding opportunities – how to engage with the HNCDI programme

This webinar is provided in association with the NAFEMS HNCDI Call which will launch in Coventry on 1 October 2024 and is open to NAFEMS members looking to access fully-funded support to adopt digital technologies.

Within the engineering community, design is, more than ever, being driven by engineering simulation. This call will offer access to specialist resources in AI, ML and related technologies to explore how the use of these digital technologies may benefit the NAFEMS community.

The following underlying themes/technologies have been identified, which we believe may provide the largest impact and be of the most interest to the engineering analysis community:

• AI and ML,
• Quantum Computing,
• Digital Twins,
• Through Life Engineering,
• Multiscale.

In alignment with UK Government National AI Strategy, the aim of this open call is to demonstrate how digital technologies such as AI, modelling and simulation can help target sustainable productivity for both design and testing as well as manufacturing, and help organisations address issues of scale and efficiency using modelling and simulation and data management.

Register for our webinar to find out what we’re all about. You’ll also have the opportunity to ask questions of our computing, big data and AI experts to explore how these technologies can improve productivity and create a competitive advantage.

S​peakers

Robert Manson-Sawko is a Senior Research Scientist at IBM Research Europe. He has worked in high-performance computing particularly with multiphase computational fluid dynamics for industrial use-cases. In recent years he has also contributed to in situ visualisation and processing, a set of flexible and scalable methods for obtaining data from large-scale simulations.

Sue Thorne is the High Performance Software Engineering Group Leader at the STFC Hartree Centre. She is an expert in HPC, numerical linear algebra and optimisation, with a particular focus on preconditioning methods. She currently leads a large portfolio of projects in collaboration with UK Atomic Energy Authority and other industrial partners.