Abstract
NEURON is the most widely used computational modeling platform in the field of neuroscience, allowing development of single cell models as well as huge networks of cells. With increase in the sizes of networks, performance and efficiency becomes a growing concern. To investigate larger networks it is necessary to have access to appropriate computational resources. It is equally imperative, and typically more economical, to extract the maximum performance from the available resources. This can be achieved by optimizing the model implementation, along with appropriate utilization of the built-in parallelization and performance enhancement features of the NEURON simulator. In the present study, we have explored, tested and benchmarked the performance benefits gained from these approaches.