by A supercomputer due to go online in April 2024 will rival the estimated rate of operations in the human brain, according to researchers in Australia. The machine, called DeepSouth, is capable of 228 trillion operations per second.
It is the world’s first supercomputer capable of simulating networks of neurons and synapses (the main biological structures that make up our nervous system) at the scale of the human brain.
DeepSouth involves an approach called neuromorphic computing, which aims to mimic the biological processes of the human brain. It will be run from the International Center for Neuromorphic Systems at the University of Western Sydney.
Our brain is the most amazing computing machine we know. By distributing its computing power to billions of tiny units (neurons) that interact through trillions of connections (synapses), the brain can compete with the world’s most powerful supercomputers, and only use the same power refrigerator lamp bulb.
Meanwhile, supercomputers usually take up a lot of space and need a lot of electrical power to run. The world’s most powerful supercomputer, the Hewlett Packard Enterprise Frontier, can do just over five billion per second. It covers 680 square meters (7,300 square feet) and requires 22.7 megawatts (MW) to run.
Our brains can perform the same number of operations per second with just 20 watts of power, while weighing just 1.3kg-1.4kg. Among other things, neuromorphic computing aims to unlock the secrets of this amazing efficiency.
Transistors at the limits
On 30 June 1945, mathematician and physicist John von Neumann described the design of a new machine, the Electronic Discrete Variable Automatic Computer (Edvac). This effectively defined the modern electronic computer as we know it.
My smartphone, the laptop I’m using to write this article have the same basic structure as the world’s most powerful supercomputer introduced by von Neumann almost 80 years ago. All of these have separate processing and memory units, where data and instructions are stored in memory and calculated by a processor.
For many years, the number of transistors on a microchip doubled every two years, something known as Moore’s Law. This allowed us to have smaller and cheaper computers.
However, transistor sizes are now approaching the atomic scale. At these tiny sizes, excessive heat generation is a problem, as is a phenomenon called quantum tunneling, which affects the operation of the transistors. This is slowing down and will eventually end transistor miniaturization.
To overcome this issue, scientists are exploring new approaches to computing, starting from the powerful computer we all hide in our heads, the human brain. Our brains do not work according to John von Neumann’s model of the computer. They do not have separate computing and memory areas.
Instead they work by connecting billions of nerve cells that communicate information in the form of electrical impulses. Information can be passed from one neuron to the next through a junction called a synapse. The organization of neurons and synapses in the brain is flexible, scalable and efficient.
So in the brain – and unlike a computer – memory and computation are controlled by the same neurons and synapses. Since the late 1980s, scientists have been studying this model with the intention of importing it to computing.
The imitation of life
Neuromorphic computers are based on complex networks of simple basic processors (which act like the brain’s neurons and synapses). The main advantage of this is that these machines are inherently “parallel”.
This means that, as with neurons and synapses, almost all processors in a computer can be operating simultaneously, communicating with each other.
In addition, because the calculations made by individual neurons and synapses are very simple compared to traditional computers, the energy consumption is orders of magnitude smaller. Although neurons are sometimes thought of as processing units, and synapses as memory units, they contribute to both processing and storage. In other words, data is already located when the calculation requires it.
This speeds up brain computing in general because there is no separation between memory and processor, which causes slowness in classical (von Neumann) machines. But it also avoids the need to perform a specific task in accessing data from a main memory component, as occurs in traditional computer systems and consumes a significant amount of energy.
The principles we have just described are the main inspiration for DeepSouth. This is not the only neuromorphic system currently active. The Human Brain Project (HBP), funded under an EU initiative, is worth mentioning. The HBP ran from 2013 to 2023, and resulted in BrainScaleS, a machine based in Heidelberg, Germany, that mimics how neurons and synapses work.
Brain scales can simulate the way a neuron “spikes”, the way an electrical impulse travels along a neuron in our brain. This would make BrainScaleS an ideal candidate for investigating the mechanisms of cognitive processes and, in the future, mechanisms underlying serious neurological and neurodegenerative diseases.
Because they are engineered to mimic an actual brain, neuromorphic computers could be the start of a revolution. By offering sustainable and affordable computing power and allowing researchers to evaluate models of neural systems, they are an ideal platform for a range of applications. They have the potential to advance our understanding of the brain and offer new approaches to artificial intelligence.
This article from The Conversation is republished under a Creative Commons license. Read the original article.
Domenico Vicinanza does not work for, consult with, or own shares or receive funding from any company or organization that would benefit from this article this, and has not disclosed any relevant connections beyond their academic appointment.
