This title appears in the Scientific Report :
2013
Supercomputers ready for use as discovery machines for neuroscience
Supercomputers ready for use as discovery machines for neuroscience
NEST is a widely used tool to simulate biological spiking neural networks [1]. The simulator is subject tocontinuous development, which is driven by the requirements of the current neuroscientific questions. Atpresent, a major part of the software development focuses on the improvement of the simula...
Saved in:
Personal Name(s): | Kunkel, Susanne (Corresponding author) |
---|---|
Schmidt, Maximilian / Eppler, Jochen Martin / Igarashi, Jun / Masumoto, Gen / Fukai, Tomoki / Ishii, Shin / Plesser, Hans Ekkehard / Morrison, Abigail / Diesmann, Markus / Helias, Moritz | |
Contributing Institute: |
Jülich Supercomputing Center; JSC Computational and Systems Neuroscience; IAS-6 Computational and Systems Neuroscience; INM-6 |
Imprint: |
2013
|
Conference: | 10th Meeting of the German Neuroscience Society, Goettingen (Germany), 2013-03-13 - 2013-03-18 |
Document Type: |
Poster |
Research Program: |
SimLab Neuroscience W2/W3 Professorinnen Programm der Helmholtzgemeinschaft Supercomputing and Modelling for the Human Brain Brain-inspired multiscale computation in neuromorphic hybrid systems The Next-Generation Integrated Simulation of Living Matter Helmholtz Alliance on Systems Biology Brain-Scale Simulations Computational Science and Mathematical Methods Signalling Pathways and Mechanisms in the Nervous System |
Publikationsportal JuSER |
NEST is a widely used tool to simulate biological spiking neural networks [1]. The simulator is subject tocontinuous development, which is driven by the requirements of the current neuroscientific questions. Atpresent, a major part of the software development focuses on the improvement of the simulator'sfundamental data structures in order to enable brain-scale simulations on supercomputers such as theBlue Gene system in Jülich and the K computer in Kobe. Based on our memory-usage model [2], weredesigned the neuronal and the connection infrastructure of NEST such that networks of 10^8 neuronsand 10^12 synapses can be simulated on the K computer [3]. These improvements reduce the memoryfootprint without compromising on the simulator's general usability and user interface. Here, we describethe recent technological advances which enable NEST to achieve high performance and good scaling ofnetwork setup and simulation on the K computer and on the Blue Gene system. We demonstrate that theusability of these machines for network simulations has become comparable to running simulations on asingle PC. |