The University of Manchester’s School of Computer Science has developed the world’s largest neuromorphic supercomputer. Developed over 10 years, the Spikking Neural Network architecture (SpiNNaker) is capable of simulating channels of the human brain, which is a complex model that consists of billions of billions of neurons.
The SpiNNaker model can also model the brain’s cortex, which has 80 million neurons and is responsible for perception, awareness, and sight. SpiNNaker is capable of completing 200 million million actions per second, making it more efficient than previous models. The project, which had cost 15 million pounds, was originally supported by the Engineering and Physical Sciences Research Council (EPSRC). More recently, it has been funded by the European Human Brain Project. The computer has finally been turned this week!
SpiNNaker works differently compared to other supercomputers, and is time and energy efficient. Instead of sending messages from one place to another independently, it sends messages in a process similar to our brains, which sends billions of pieces of information simultaneously to different destinations.
Steve Furber, one of the founders of the Human Brain Project, said, “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 modelling applications, and we have now achieved it, which is fantastic.”
The creators of this computer wanted to model a billion biological neurons in real time and they have gotten closer to achieving their objective. They have achieved 1% of the human brain’s processing power, which has approximately 100 billion neurons, each linked closely by quadrillions of synapses.
The supercomputer could not only bring breakthroughs in our understanding of how the human brain works, but also in research around neurological diseases. Basel Ganglia, a region of the brain that is affected by Parkinson’ disease has been simulated. SpiNNaker has recently even been used to control robots like the Spomnibot. The robot uses the SpiNNaker system to make decisions like navigating to certain objects and ignoring others.
Professor Furber added: “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 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.”
Supercomputers had been introduced in the 1960s and the The United States Department of Energy’s Oak Ridge National Laboratory (ORNL) supercomputer “Summit” recently was considered as the fastest supercomputer. It could be used for research in artificial intelligence, energy, advanced materials, cancer, Alzheimers’ disease, sustainability and much more.
The SpiNNaker will be used for unlocking the “secrets of the brain” and are part of a group that have been described as “…a race to all human knowledge – a race to understand everything.”