Ctrl K

FLAME GPU

FLAME GPU is a GPU accelerated simulator for domain independent complex systems simulations. FLAME GPU provides a mapping between a simple description of an agent and its interactions into optimised GPU code. The software abstracts the details and complexity of the GPU away from modellers.

4
contributors

Description

  • Provides the ability for agent based modellers to focus on specifying agent behaviour and run simulations without explicit understanding of CUDA programming or GPU optimisation strategies.
  • Simulation performance of an agent based model is significantly increased in comparison with desktop CPU alternatives. This allows simulation of far larger model sizes (or ensemble experiments) with high performance at a fraction of the compute/energy cost of grid based alternatives.
  • Massive agent populations can be visualised in real time as agent data is already located on the GPU hardware.
  • Utilise Python to specify models.
Logo of FLAME GPU
Keywords
GPU
High performance computing
Multi-scale & multi model simulations
python
Visualization
License
</>Source code

Participating organisations

University of Sheffield

Contributors

Paul Richmond
Paul Richmond
Project Lead
University of Sheffield
PH
Peter Heywood
RC
Robert Chisholm
ML
Matthew Leach