Spiking Network Computer Architecture, nicknamed SpiNNaker to simplify its name, is a supercomputer that has a million processing cores and 1,200 interconnected circuit cards, all working together to simulate the functioning of a human brain, capable of containing 200 quadrilioni of simultaneous actions. We are talking in numbers of as much as 200,000,000,000,000,000,000,000,000,000,000 actions that occur simultaneously. This huge number makes this computer the best we have so far, in terms of machines that simulate our brains making it the largest neuromorphic computer in the world.


This machine, owned by the University of Manchester, has required for its development not less than 20 years, the project was born in fact, in 1998, thanks to the EPSRC funds and, subsequently, to those of the Human Brain Project.  A neuromorphic computer not only “thinks” like a human brain, but can also create models of neurons and simulate more in real time, faster than any other computer can do. With SpiNNaker’s current processing capabilities, researchers believe it will be possible to create detailed models of our brain, which is extremely complex and not completely exploited by us.

And why is SpiNNaker ahead of other supercomputers of this type, which may even have more processors?

Apparently conventional supercomputers have much less optimized connectivity mechanisms for real-time brain modeling. According to Steve Furber, a professor of computer engineering at the University of Manchester, SpiNNaker is able to model neural networks with more real-time biological realism, more than any other machine ever built. However, SpiNNaker is still far from behaving like a real human brain, unfortunately. Furber explains that


“its main task is to support partial brain models[…] and even with one million processors, we can only approach the 1% of the capacity of the human brain and this on the basis of many simplifying hypotheses”




But SpiNNaker has already been used to mimic an area of the brain affected by the Parkinson’s disease and can be used as a tool to study neurological disorders in an even more technologically advanced way. SpiNNaker emulates the parallel communication architecture of the brain that simultaneously sends billions of small amounts of information to thousands of different destinations, much like neurons transmit chemical and electrical signals through the brain. Always according to Steve Furber SpiNNaker completely revises the way conventional computers work. Essentially, a machine was built that works more like a brain than a traditional computer, which is extremely exciting.

Your Remaining Votes (within 24hrs) : 10 of 10
8 votes, average: 4.63 out of 58 votes, average: 4.63 out of 58 votes, average: 4.63 out of 58 votes, average: 4.63 out of 58 votes, average: 4.63 out of 5 (8 votes, average: 4.63 out of 5)
You need to be a registered member to rate this.
(1170 total tokens earned)