Scientists just switched on the largest neuromorphic supercomputer in the world.
With one million processor cores, the Spiking Neural Network Architecture, or SpiNNaker, is now the world’s largest neuromorphic supercomputer. On November 2nd, the developing researchers from the University of Manchester turned on the supercomputer.
According to the UM research team, each chip on the SpiNNaker has around 100 million parts. This enables the computer to perform over 200 trillion operations per second. Unlike other neural network architectures, the SpiNNaker was designed to mimic the biological neural networks in a human brain by using Army9 processors.
The building block of the neural machine, the SpiNNaker SoC is reportedly comprised of Army968 chips integrated with a “network-on-chip”. This chip is responsible for performing the packet-switched asynchronous communications of the system.
The Neuromorphic Supercomputer
SpiNNaker’s digital assets work collectively to help it mimic the human brain’s capabilities. The supercomputer mirrors the way neurons communicate with each other. It then sends billions of pieces of information in parallel to different core processors instead of relying on virtual models of the brain. This process allows the machine to copy the way biological neurons process information.
“SpiNNaker completely re-thinks the way conventional computers work. We’ve essentially created a machine that works more like a brain than a traditional computer, which is extremely exciting. The ultimate objective for the project has always been a million cores in a single computer for real-time brain modeling applications, and we have now achieved it, which is fantastic,” Steve Furber, a professor of Computer Engineering at the University of Manchester, said.
According to Furber, the SpiNNaker could help the neuroscientist community better understand how the brain functions. It could also help the robotics industry integrate the brain power needed to control their robots into their calculations.
“Neuroscientists can now use SpiNNaker to help unlock some of the secrets of how the human brain works by running unprecedentedly large-scale simulations. It also works as a real-time neural simulator that allows roboticists to design large-scale neural networks into mobile robots so they can walk, talk and move with flexibility and low power,” Furber went on to say.