Neurogrid - Brains in Silicon, Stanford University

Last updated: February 20, 2013

Two complementary objectives: To use existing knowledge of brain function in designing
an affordable supercomputer�one that can itself serve as a tool to investigate brain function
- feeding back and contributing to a fundamental, biological understanding of how the brain works.

Neurogrid provides an affordable option for brain simulations.
It uses analog computation to emulate ion-channel activity
and uses digital communication to softwire synaptic connections.

Neurogrid simulates one million neurons with two subcellular compartments each.
varying the compartments' electrical coupling replicates the firing patterns of
various pyramidal-cell types. Capturing these behaviors using the smallest number
of compartments minimizes the number of distinct ion-channel populations that
need to be simulated, thereby maximizing the number of neurons a model can have.

With sixteen 12x14 sq-mm chips (Neurocores) assembled on a 6.5x7.5 sq-in circuit board,
Neurogrid can model a slab of cortex with up to 16x256x256 neurons�over a million.
The chips are interconnected in a binary tree by 80M spike/sec links.
An on-chip RAM (in each Neurocore) and an off-chip RAM (on a daughterboard)
softwire vertical and horizontcal cortical connections, respectively.
Built in 2009 (maybe also earlier)

Neurogrid simulates a million neurons connected by billions of synapses in real-time,
rivaling Blue Gene's performance while consuming a 100,000 times less energy�five watts
instead of a megawatt

Contents    (hide)

Latest news / current status

  • Neurogrid possibly to be used with the SPAUN model to achieve real-time simulation.

Neuromorphic hardware


Simulation experiments


Future directions

Neurogrid possibly to be used with the SPAUN model.

Kwabena Boahen
New project: Fully implantable and programmable spike-based codecs for neuroprosthetics
Two neuromorphic chips, one implanted into motor cortex, wirelessly controls robotic arm,
another into somato sensory cortex, receives sensory data from arm
Project Start Date:	1-SEP-2011
Project End Date:	31-MAY-2016
Total funding: $895,817

People involved




Most recent papers: