Arbor: A morphologically detailed neural network simulator for modern high performance computer architectures
| Autor: | Cumming, B., Yates, S., Klijn, Wouter, Peyser, Alexander, Karakasis, V., Perez, I. |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2017 |
| Zdroj: | Neuroscience 2017: Society for Neuroscience, Washington, DC, United States, 2017-11-15-2017-11-15 HBP summit 2017, Glasgow, England, 2017-10-17-2017-10-20 |
| Popis: | Arbor is a new multicompartment neural network simulator currently under development as a collaboration between the Simulation Lab Neuroscience at the Forschungszentrum Jülich, the Barcelona Supercomputing Center and the Swiss National Supercomputing Center. Arbor will enable new scales and classes of morphologically detailed neuronal network simulations on current and future supercomputing architectures.A number of "many-core" architectures such as GPU and Intel Xeon Phi based systems are currently available. To optimally use these emerging architecture new approaches in software development and algorithm design are needed. Arbor is being written specifically with performance for this hardware in mind (Fig. 1); it aims to be a flexible platform for neural network simulation while keeping interoperability with models and workflows developed for NEST and NEURON.The improvements in performance and flexibility in themselves will enable a variety of novel experiments, but the design is not yet finalized, and is driven by the requirements of the neuroscientific community. The prototype is open source (https://github.com/eth-cscs/nestmc-proto , http://usertest.cscs.ch/nestmc/) and we invite you to have a look. We are interested in your ideas for features which will make new science possible: we ask you to think outside of the box and build this next generation neurosimulator together with us.Which directions do you want us to go in?Simulate large morphological detailed networks for longer time scales: Study slow developing phenomena.Reduce the time to solution: Perform more repeat experiments for increased statistical power.Create high performance interfaces with other software: Perform online statistical analysis and visualization of your running models, study the brain at multiple scales with specialized tools, or embed detailed networks in physically modeled animals.Optimize dynamic structures for models with time-varying number of neurons, synapses and compartments: simulate neuronal development, healing after injury and age related neuronal degeneration. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|