Lecturer, Department of Computer Science, Brunel University London
I am interested in simulations of blood flow, refugee movements, materials, and star clusters. I also focus on multiscale computing, high-performance computing, and distributed computing.
I am a Lecturer in Simulation and Modelling and I use and investigate large scale simulations in high-performance computing environments. In doing so, I look for generalized ways to make large scale simulations simpler, more relevant, more flexible and more efficient. I have a particular preference for multiscale or multi-model simulations, where effective combination of existing codes is the method of choice for addressing new science questions.
I am responsible for improving and maintaining the FabSim automation toolkit and the MPWide communication library, which we have used to couple different simulation codes and to parallelize simulations across multiple supercomputers. I worked on MUSCLE 2; a toolkit designed to provide structured code coupling based on a formalized multiscale modelling framework.
I am working with a wide range of applications, including methods such as lattice-Boltzmann, stellar and cosmological N-body, agent-based simulations of refugees and molecular dynamics. I finished my PhD in 2010, where I ran large cosmological simulations geographically distributed across up to four supercomputers. I have also worked on a variety of performance optimization projects, and among other things enabled lattice-Boltzmann codes to run efficiently using up to 262,144 cores. More information about me is available on my web page.