First launched the most powerful supercomputer simulating the human brain

Today supercomputers are used for a wide range of tasks from various mathematical calculations and manipulation of data sets to simulation of pharmaceutical compounds and artificial intelligence. However, there are computers aimed at maximally accurate reproduction of the “architecture” of the human brain. And the most powerful of such neuromorphic supercomputer was recently launched.

This computer is called a Spiking Neural Network Architecture (SpiNNaker) and it can carry more than 200 million operations per second, and each chip has 100 million parts. The machine is assembled at the School of computer science the University of Manchester and the project has been spent 15 million pounds. But this amount is not as impressive as the time taken for development: 20 years have studied the concept of a supercomputer and it took 10 years to build.


Chip supercomputer SpiNNaker

SpiNNaker can in real time to simulate the behavior of human neurons and it is unique in that unlike traditional computers, it does not transmit large amounts of information from point a to point B, and sends data simultaneously in thousands of different directions. Work this way the cells in our brain. As stated by one of the ideologists of the project, Professor of computer science Steve Furber,

“SpiNNaker is changing the way conventional computers. In fact, we have created a machine that works more like a brain than a traditional computer, which is extremely interesting. The ultimate goal of our project has always been to use a billion cores in one computer, which simulated the functioning of the brain and today we achieved that goal. And it’s just fantastic.”

It should be noted that even despite the fact that “true artificial brain” is earned, the number of cores that it is now, is a model of the brain, which is the equivalent of 1% of the human. However, now this neuromorphic supercomputer will help to understand the functioning of our Central nervous system, and will enable large-scale simulations are not available on traditional machines.

This and other news you can discuss in our chat in Telegram.


Date:

by