This short-course, developed January 2024, is my first step in developing a community of practice around the topic of software performance. The aim is to make performance optimisation techniques accessible to programmers at all levels of skill. The initial development of this course was funded by the University of Sheffield, linked to the launch of the Stanage HPC cluster. This course was first announced externally in September 2024 at RSECon24.
This project proposes an open cloud-based platform to support decision making in the clinical management of two paediatric cancers, Neuroblastoma (NB), the most frequent solid cancer of early childhood, and the Diffuse Intrinsic Pontine Glioma (DIPG) the leading cause of brain tumour-related death in children.
At Sheffield we are developing an agent-based cellular scale model of Neuroblastoma, which is being accelerated with GPU computation.
This project is funded as part of the EU's Horizon 2020 scheme.
My research concerns improvement to the Uniform Spatial Partitioning data structure, used to provide Fixed Radius Near Neighbours search within complex systems simulations. This permits thousands to millions of entities to simultaneous survey their local environment during simulations. Due to the highly scattered memory accesses and non-uniform distributions experienced, it is often one of the most costly processes during models such as pedestrian simulations.
There is a constant pressure from industry to deliver larger and more detailed simulations, in order to provide these improvements to the performance of the underlying algorithms driving the models must be achieved.
This research project delivered a proof of concept, demonstrating how three independent transport models could be integrated to model the passenger life cycle.