high performance computing on graphics processing units: hgpu.org

There is a huge computational potential in unconventional computing paradigms such as reaction-diffusion chemistry. The main problem with unconventional systems is the inherent difficulty in programming them. By extending the computational application of reaction-diffusion systems, this problem may be alleviated, as every new application allows for another method of approaching problems. With the central nervous system, biology has evolved an innately parallelised architecture. Modelling neural networks, therefore, should allow for the parallelism present in reaction-diffusion systems to be utilised to full effect.