A supercomputer simulating the human brain will be launched in 2024

Mondo Science Updated on 2024-01-29

**:Technology**.

According to the British "New Scientist" weekly **12**, a supercomputer capable of fully simulating the synapses of the human brain will be launched in Australia next year. The neuromorphic supercomputer, called Deep South, can perform 228 trillion synaptic operations per second, which is comparable to the number of synaptic operations scientists estimate in the human brain, and will help understand how the human brain processes large amounts of information while consuming relatively little energy.

The research team noted that similar neuromorphic computers have been available before, but Shennan will be the largest yet. Deep South is manufactured by the International Center for Neuromorphic Systems in Sydney, Australia, in collaboration with Intel and Dell. Unlike ordinary computers, Deep South's hardware chips enable spike neural networks to model the way synapses process brain information.

Andrew van Shaick, project leader at Deep South, noted that this will be the first time they have simulated the activity of a spike neural network the size of a human brain in real time. While Deep South isn't as powerful as existing supercomputers, it will help advance the understanding of neuromorphic computing and the biological brain, leading to better insights into how the brain works.

Supercomputers are big consumers of energy, and the human brain consumes no more energy than a light bulb. This difference is partly due to the different ways in which data is handled: while a traditional computer performs an operation, data is constantly moving between the processor and the memory;Whereas, the neuromorphic architecture performs many operations in parallel, greatly reducing the movement of data. Since data movement is one of the most energy-intensive parts, neuromorphic architecture can significantly reduce energy consumption. In addition, spike neural networks are event-driven, which means that neuromorphic systems only react to changes in input, rather than running continuously in the background like traditional computers, further reducing energy consumption.

The researchers say Deep South will boost neuroscience research and pave the way for greater computing energy efficiency. If the technology can be miniaturized, it will improve the autonomy of drones and robots.

Related Pages