Code that writes code: smart tools for high performance, high fidelity modelling
From seismic imaging to drug discovery, high performance simulation and modelling drives business performance and competitive advantage across industries. But realizing this potential requires a specialist combination of domain-specific, mathematics and programming skills. The “skills crunch” for high performance computing slows innovation, drives up costs and acts as a barrier-to-entry for SMEs.
But what if the required knowledge could be packaged up and made accessible to every user? Smart tools would free users to focus on the problems they want to solve - rather than the distracting challenges of programming a high performance computer. Imperial College is leading the world in the development of next-generation software tools that will make the benefits of high-performance modelling accessible to anyone.
I'm curious about..."the democratisation of tools that will allow anyone, in principle, to capitalize on the power of the fastest computers on the planet"Gerard Gorman
Gerard Gorman is Reader in Computational Science in the Department of Earth Science and Associate Director of the Centre for Computational Methods in Science and Engineering.
Gerard is an expert in the development of software-based algorithms to complex dynamics problems such as fluid-structure interactions, exploiting novel techniques such as anisotropic adaptive mesh methods. He is a major contributor to PRISM (Platform for Research In Simulation Methods).
Applied work focuses on a range of topics, including:
- Geophysical fluid dynamics
- Seismic imaging, through the Intel Parallel Computing Centre for Open Performance Portable Seismic Imaging (OPESCI)
- Renewable energy
Foresight and futures work
The Future Of Simulation, Design, Optimisation And Prediction For The Digital Economy