Logo of the Helmholtz Platform for Research Software Engineering

NEST Simulator

DOI

NEST is a simulator for spiking neuronal networks. It is used mostly in computational neuroscience to model and study behaviour of large networks of neurons. The models describe single neuron and synapse behaviour and their connections. Different mechanisms of plasticity can be used to investigate artificial learning and help to shed light on the fundamental principles of how the brain works.

NEST is a well tested and efficient tool that works on your laptop and also on the world’s largest supercomputers. It is fast and memory efficient, making best use of your multi-core computer or compute cluster. Whether you want to work on a small laptop or the largest supercomputers, NEST can seamlessly scale to your needs.

A NEST simulation works like a physiological experiment but inside a computer. Perform your own experiments using assemblies of virutal neurons and different stimulating and recording devices. It comes loaded with numerous state-of-the art neuron models. Textbook standards like integrate-and-fire and Hodgkin-Huxley type models are available alongside high quality implementations of models published by the neuroscience community. NESTML provides a framework to create models without the need for detailed programming knowledge, and makes creating your own neuron and synapse models a breeze! Visit NESTML to find out more.

NEST offers convenient and efficient commands to define and connect large networks, ranging from algorithmically determined connections to data-driven connectivity. Create connections between neurons using numerous synapse models from STDP to gap junctions.

The simulator is developed and continuously improved by the NEST community around the globe to stay up-to-date and fit for the latest research questions. NEST developers are using continuous-integration based workflows in order to maintain high code quality standards for correct and reproducible simulations.

NEST has fostered a large community of experienced developers and amazing users, who actively contribute to the project. Our community extends to related projects, like the teaching tool NEST Desktop, cross-simulator languages like PyNN and neural activity analysis tools like Elephant.

With the combination of AI and neuromorphic computing, both sides can advance from findings from both worlds. Connect your NEST simulation to neurorobotics and add the “eyes” and “muscles” to complete the sensor-motor loop and tackle the problem of embodiment.

Activities

  • Participated in continuous benchmarking hackathon
  • Presented documentation related talk to HiRSE Seminar Series
  • Improved CI chain with new RSE best practice tools