Neuron splitting in compute-bound parallel network simulations enables runtime scaling with twice as many processors

TitleNeuron splitting in compute-bound parallel network simulations enables runtime scaling with twice as many processors
Publication TypeJournal Article
AuthorsHines, Michael L., Eichner Hubert, and Schürmann Felix
Full Text

Preprint available as splitcell.pdf 
Load balance is important for maximizing speedup when simulating neural networks on parallel hardware. With NEURON, load balance can be achieved by splitting cells into subtrees that are solved on different processors with no change in accuracy, stability, or computational effort; interprocessor communication costs are minimal.