(The Center Square) – Sandia National Laboratories researchers are welcoming the arrival of a brain-based computing system, known as Hala Point. The neuromorphic brain features pulsing neurons by the millions, cutting edge chips and could rewrite the way we compute from the design of nuclear weapons to autonomous vehicles on your street. It is “20 times faster than a human brain,” according to Intel which hasn’t disclosed the cost of the system..

“Packed with a staggering 1.15 billion artificial neurons — believed to be the biggest brain-based computing system in the world — and cleverly confined within a container roughly the size of a microwave oven, this technological marvel had made its journey to Albuquerque, New Mexico, from its birthplace at Intel Corp. in Portland, Oregon,” Sandia Labs revealed.

The system will provide Sandia and the National Nuclear Security Administration research teams with the tools to realize large-scale brain-based computing.

“At a smaller scale, the neuromorphic method has already demonstrated greater speed, accuracy, and lower energy costs than conventional computing in several labs, including Sandia,” a release said.

The new system is 10 times faster, and 15 times denser and has increased from 128,000 circuits on a single chip to one million, compared with the system of 50 million artificial neurons known as Pohoiki Springs that Sandia received from Intel in 2021.

“We believe this new level of experimentation — the start, we hope, of large-scale neuromorphic computing — will help create a brain-based system with unrivaled ability to process, respond to, and learn from real-life data,” Sandia lead researcher Craig Vineyard said.

The two systems use two generations of research chips, which are named Loihi 1 and 2, after the youngest volcano in the Hawaiian Islands. The latest computing system uses 1,152 Loihi 2 research processors.

Though his group is focused on problems involving large-scale physics, chemistry, and the environment, Vineyard said the technique could be disruptive at multiple scales.

A spokesman for Sandia Labs also told The Center Square that the development “can be a foundation for AI, extending its capabilities.”

“On the one hand, we’re looking at science codes, physics computations,” Vineyard said. “Can we model large processes in more detail? What about device design or climate models with better-defined resolution?”

Vineyard said this means creating applications capable of using the full system.

Vineyard thinks the technology will help create smarter soldier gear, better analysis of intelligence operations, better border security, and quicker response to earthquakes. He also thinks it will help with faster medical diagnosis and less expensive drug discovery.

“We might soon see self-driving cars with neuromorphic technology, lane detection, cell phones with voice recognition, smarter watches, and refrigerators, more detailed home security systems,” Vineyard said. “Did the cat run by, or is someone in your house?”

The neuromorphic process saves computing time and energy because it only electrically pulsates when a “synapse in a complex circuit has absorbed enough charge to produce an electrical spike,” the release said.

“This process discards useless information — that which doesn’t spike — instead of storing it in distant locations and revisiting it in every calculation,” the release said. In this manner, neuromorphic computing operates like the brain does, with subgroups of active neurons arranged in parallel circuits, tapped for information as needed, rather than the sequential instructions and remote memory storage involving every possible unit that characterizes mainstream computing.”

Sandia researcher Brad Aimone compared the technology to a human brain.

“One of the main differences between brain-like computing and regular computers we use today — in both our brains and in neuromorphic computing — is that the computation is spread over many neurons in parallel, rather than long processes in series that are an inescapable part of conventional computing,” Aimone said. “As a result, the more neurons we have in a neuromorphic system, the more complex a calculation we can perform. We see this in real brains. Even the smallest mammal brains have tens of millions of neurons; our brains have around 80 billion. We see it in today’s AI algorithms. Bigger is far better.”

The work is funded by The NNSA’s Advanced Simulation and Computing program.

“The NNSA is a semiautonomous DOE agency responsible for the management and security of the nation’s nuclear weapons, nuclear nonproliferation, and naval reactor programs, as well as responding to nuclear and radiological emergencies in the U.S. and abroad,” the release said.