This title appears in the Scientific Report : 2014 

Integrating brain structure and dynamics on supercomputers
van Albada, Sacha (Corresponding Author)
Kunkel, Susanne / Morrison, Abigail / Diesmann, Markus
Jülich Supercomputing Center; JSC
Theoretical Neuroscience; IAS-6
Computational and Systems Neuroscience; INM-6
Brain-inspired Computing
Cham Heidelberg New York Dordrecht London Springer 2014
978-3-319-12083-6 (print)
978-3-319-12084-3 (electronic)
1st International Workshop on Brain-inspired Computing, Cetraro (Italy), 2013-07-08 - 2013-07-11
Contribution to a book
Contribution to a conference proceedings
Helmholtz Alliance on Systems Biology
The Next-Generation Integrated Simulation of Living Matter
Supercomputing and Modelling for the Human Brain
Brain-Scale Simulations
SimLab Neuroscience
The Human Brain Project
W2/W3 Professorinnen Programm der Helmholtzgemeinschaft
Brain-inspired multiscale computation in neuromorphic hybrid systems
Theory, modelling and simulation
Computational Science and Mathematical Methods
Signalling Pathways and Mechanisms in the Nervous System
Lecture Notes in Computer Science 8603
Please use the identifier: in citations.
Large-scale simulations of neuronal networks provide a unique view onto brain dynamics, complementing experiments, small-scale simulations, and theory. They enable the investigation of integrative models to arrive at a multi-scale picture of brain dynamics relating macroscopic imaging measures to the microscopic dynamics. Recent years have seen rapid development of the necessary simulation technology. We give an overview of design features of the NEural Simulation Tool (NEST) that enable simulations of spiking point neurons to be scaled to hundreds of thousands of processors. The performance of supercomputing applications is traditionally assessed using scalability plots. We discuss reasons why such measures should be interpreted with care in the context of neural network simulations. The scalability of neural network simulations on available supercomputers is limited by memory constraints rather than computational speed. This calls for future generations of supercomputers that are more attuned to the requirements of memory-intensive neuroscientific applications.