The Neuroscience Gateway (NSG) is an open project funded by the National Science Foundation and developed by researchers from UCSD and Yale University. The NSG (www.nsgportal.org) is a software infrastructure that allows computational neuroscience researchers and students to easily utilize HPC and data resources to make neuroscience-specific simulations. It allows uploading of models, specifying parameters for running simulations, and retrieving and storing output data. The NSG provides neuronal software such as NEURON, GENESIS3, MOOSE, NEST, and PyNN. Computational neuroscience has seen tremendous growth in the recent years. This is evident from the large number of publications, in prestigious neuroscience journals, that are more and more based on modeling in the field of neuroscience. This has motivated development of parallel simulation environments such as NEURON, GENESIS3, MOOSE, NEST, and PyNN. During this same time there has been significant development of Cyberinfrastructure resources consisting of High Performance Computing (HPC) machines, fast networks, next generation data and storage technologies and software that can create and manage complex workflow, automatically run parallel simulations, and handle data/output retrieval/transfer. The more complex neuroscience problems, which involve network models, optimization or exploration of high dimensional parameter spaces etc., require access to HPC machines and data storage. Access to national scale CI/HPC can help broaden the base of students and researchers, who can attempt to learn and use computational neuroscience even if a first class HPC center does not exist at their own university or college. Yet accessing and using HPC resources remain difficult for neuro-scientists due to the need to (i) write convincing allocation proposals, (ii) understand complex HPC architectures, (iii) master complex OS/software environment, (iv) optimally install neuronal simulation software, (v) learn policies and batch system environments, (vi) manage data transfer and retrieval, and (vii) understand remote authentication process. This project involves building a Neuroscience Gateway (NSG) which will simplify the CI/HPC environment and reduce the barrier to access and use HPC resources for neuro-scientists. The proposed NSG gateway will provide optimally installed parallel neuronal software (NEURON, GENESIS3, MOOSE, NEST and PyNN) on HPC resources of XSEDE. We expect that, following the success of other science gateways, this NSG will allow large number of neuro-scientists easily and efficiently use these parallel neuronal software on XSEDE HPC resources and enable, enhance and broaden computational neuroscience research and education.
Cite this work
Researchers should cite this work as follows: