Header Ads

A Super Computer Mimics Brain Cells, Reaches 1% Of Capacity

Using a Japanese super-computer, a team of scientists have carried out the largest ever imitation of the way a brain's cells connect with each other. This will pave the way for a better understanding of the extremely complex human brain, they say.

Researchers from the RIKEN, a Japanese research body, the Okinawa Institute of Science & Technology (OIST) in Japan and the Forschungszentrum Julich, a research institute in Germany, used the K-computer in Julich to carry out the neuronal network simulation.

The brain consists of some 200 billion nerve cells, also called neurons. These are linked to each other by trillions of connections called synapses. Small electrical impulses are fired across the neurons through synapses, each of which contains an estimated 1000 switches for routing the message. The total network runs into hundreds of trillions of pathways. Just the topmost layer of the brain, the cerebral cortex has an estimated 125 trillion synapses, according to research by Stephen Smith, a professor at Stanford University.

The latest simulation managed to create a virtual or electronic neuronal network of 1.73 billion nerve cells connected by 10.4 trillion synapses. For this feat, it used a open-source software called NEST and 82,944 processors of the K computer. The process took 40 minutes, to complete the simulation of 1 second of neuronal network activity in real, biological, time.

Although the simulated network is huge, it only represents 1% of the neuronal network in the brain. The nerve cells were randomly connected and the simulation itself was not supposed to provide new insight into the brain - the purpose of the endeavor was to test the limits of the simulation technology developed in the project and the capabilities of K. In the process, the researchers gathered invaluable experience that will guide them in the construction of novel simulation software.

The team was led by Markus Diesmann in collaboration with Abigail Morrison both now with the Institute of Neuroscience and Medicine at Julich.

Simulating a large neuronal network and a process like learning requires large amounts of computing memory. Synapses, the structures at the interface between two neurons, are constantly modified by neuronal interaction and simulators need to allow for these modifications.

More important than the number of neurons in the simulated network is the fact that during the simulation each synapse between excitatory neurons was supplied with 24 bytes of memory. This enabled an accurate mathematical description of the network. In total, the simulator coordinated the use of about 1 petabyte of main memory, which corresponds to the aggregated memory of 250,000 PCs.

"If peta-scale computers like the K Computer are capable of representing 1% of the network of a human brain today, then we know that simulating the whole brain at the level of the individual nerve cell and its synapses will be possible with exa-scale computers hopefully available within the next decade," explains Diesmann. One 'peta' is a thousand trillion and one 'exa' is a thousand 'peta'.

The achievement on K computer is encouraging news for the Human Brain Project (HBP) of the European Union, scheduled to start this October. The central supercomputer for this project will be based at the Forschungszentrum Julich.
Powered by Blogger.