Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Large neural network simulations on multiple hardware platforms

P Hammarlund1, O Ekeberg

  • 1Studies of Artificial Neural Systems, Department of Numerical Analysis and Computing Science, Royal Institute of Technology, Stockholm, Sweden.

Journal of Computational Neuroscience
|January 7, 1999
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Posttraumatic stress disorder symptoms in adults with congenital heart disease (ACHD) during the COVID-19 pandemic in Norway.

International journal of cardiology. Congenital heart disease·2025
Same author

Artificial intelligence for direct-to-physician reporting of ambulatory electrocardiography.

Nature medicine·2025
Same author

Prevalence of suicide attempt and associations with deliberate self-harm, mental health problems, drug misuse and traumatic experiences - a cross sectional survey of the Norwegian population.

BMC psychiatry·2024
Same author

Neural and non-neural related properties in the spastic wrist flexors: An optimization study.

Medical engineering & physics·2017
Same author

Healthcare utilisation, induced labour and caesarean section in the pregnancy after stillbirth: a prospective study.

BJOG : an international journal of obstetrics and gynaecology·2017
Same author

Lower suicide intention in patients with personality disorders admitted for deliberate self-poisoning than in patients with other diagnoses.

Annals of general psychiatry·2017
Same journal

Hierarchical learning creates invariant schema within plastic neural networks.

Journal of computational neuroscience·2026
Same journal

Intrinsic chaos control in cortical circuits: A minimal E-I-M rate model for primary visual cortex.

Journal of computational neuroscience·2026
Same journal

Modeling developmental spiking behavior driven by ionic current dynamics of mouse and human inner hair cells using a calcium-enhanced Izhikevich framework.

Journal of computational neuroscience·2026
Same journal

A biophysically grounded model of glutamatergic synaptic transmission integrating glutamate transport, receptor kinetics, and electrotonic effects.

Journal of computational neuroscience·2026
Same journal

When can neuronal activity-dependent homeostatic plasticity maintain circuit-level properties?

Journal of computational neuroscience·2026
Same journal

A charge conservative finite volume discretization of the Hodgkin-Huxley model.

Journal of computational neuroscience·2026
See all related articles

This study presents techniques for efficient neural network simulation across diverse computer architectures. A new simulation library, SPLIT, was developed to enable portable and extendable code for large-scale neuronal modeling.

Area of Science:

  • Computational Neuroscience
  • Computer Science

Background:

  • Simulating large-scale neural networks requires careful consideration of computer architecture.
  • Existing simulators may not be optimized for diverse hardware.

Purpose of the Study:

  • To present techniques for implementing efficient neural network simulators on various computer architectures.
  • To adapt existing simulation tools and develop a new library for enhanced portability and performance.

Main Methods:

  • Characterization of neuronal simulation tasks and target computer architectures.
  • Adaptation of the SWIM simulator to vector computers and multiprocessor workstations.
  • Development of the SPLIT simulation library, separating architecture-specific concerns.

Main Results:

Related Experiment Videos

  • Identified potential bottlenecks in simulating large neural networks.
  • Successfully adapted an existing simulator (SWIM) to different architectures.
  • Implemented the SPLIT library for efficient and portable large-scale network simulation.

Conclusions:

  • Separating architecture considerations from neuronal models allows for portable and extendable simulation code.
  • The SPLIT library facilitates efficient simulation of large neural networks on multiple architectures.
  • Optimized simulation techniques are crucial for advancing computational neuroscience research.